The overlaps of DCIM with inventory, asset and config management

A regular reader of the PAOSS blog recently wrote, “I follow with passion your blog,latest post about Inventory are great [Ed. the reader is talking about this post about LNI and PNI and this one about Inventory vs Asset vs CMDB Management]. I ask you if possible have a post on Inside Plant vs Outside Plant vs Virtual network creation… we usually use CAD based tool for Inside Plant design both for TLC equipment, cabling, cross connection, Distribution Frame, rooms, virtual rooms, rows structure,etc but also for power, conditioning, lighfiring,etc. We also use Network Inventory for Datacenter and server farm modelling.Outside Plant typically deals with GIS tool for cabling infrastructure. And now also virtualizzation of Network is coming with NFV and SDN. What do you think about?”

Great question.

In the post about Inventory vs Asset vs CMDB, we used the following Venn Diagram:

Unfortunately, there’s another circle that’s not shown on this diagram, but should be – the DCIM (Data Centre Infrastructure Management) circle. The overlaps between OSS and DCIM partially answer the questions above. We wrote a 5 part series on DCIM back in 2014 (part one, two, three, four, five), so perhaps it’s time for a re-visit.

The last of those five posts even included another Venn Diagram, as follows:

OSS, DCIM, ITSM Venn Diagram

Data Centre Infrastructure Management (DCIM) shares much of its DNA with OSS, but also has a number of unique differences.

Similarities:

  • IT and network device / inventory management
  • CSPs and Data Centres tend to have many Enterprise customers, and therefore a need to align with their IT service and life-cycle management (ITIL / ITSM) methodologies
  • Electronic data collection and storage to support fulfillment and assurance workflows
  • Analytics and operational decision support
  • Planning and design tools
  • Predictive modelling
  • Process and change management
  • Capacity planning, resource allocation and provisioning

Differences (ie what Data Centres have that traditional CSP networks don’t):

  • Facilities / Building Management Systems (FMS/BMS)
  • Energy / Power management
  • Environment and heat management (HVAC) including management of hot/cold zones
  • Data Centres tend to have less outside plant or inter-site connectivity* (ie most power and network connectivity tends to reside within the Data Centres)
  • However, Data Centre cable management have some slight differences. Network links are more likely to be managed within 3D spatial systems (x, y and height) if at all, rather than the 2D (x and y coordinates) typically plotted by most OSS inventory via GIS (Geographical Information Systems) or CAD (Computer Aided Design) drawings. Data Centre cables tend to be run in spatially-dense above-rack or below-floor trayways. By comparison, cables between sites tends to be less dense and at a fairly consistent height (eg a standard depth underground or a standard height when mounted on towers/poles aboveground)
  • Alternatively, DCs may manage spatial infrastructure through naming conventions such as rooms, rack-rows, racks, rack-position rather than 3D spatial systems
  • Data Centres have traditionally had a higher proportion of virtualised assets than traditional CSPs, although that is now changing with the operator network embracing network virtualisation

 

So let’s now look at how it “might” all hang together (noting that each company is likely to be different depending on their systems and processes):

  • DCIM manages facilities, building, power / PLCs and heating/cooling/HVAC
  • PNI manages physical connectivity (between sites and within the DC) as it can generally manage connectivity to physical ports on patch-panels / frames and physical devices (eg switches and routers) inside the DC. PNI also handles splicing and patching. PNI tools can generally also manage power cabling, although not everyone uses PNI for this
  • LNI (in conjunction with EMS [Element Management Systems] and virtual resource managers) will tend to manage the virtual / logical networks including resource management and orchestration
  • LNI will also tend to provide topological views of the network (often point-to-point links between physical/logical ports rather than the cable routes shown in PNI). LNI may also potentially include rack layouts and other forms of network visualisation. However, LNI tends to only partially show spatial presentation of the data (eg physical locations of “circuit” end-points rather than spatial location of all racks and equipment in 3D)
  • Related compute / storage infrastructure could be managed by DCIM, LNI, VIM, etc
  • And any of this could be cross-referenced as assets in the Asset Management System and/or Configuration Management Database (CMDB)

I can see that CAD might still be required for trayway, HVAC ducting, etc because PNI isn’t really designed with this in mind in 3D. 

Having said that, I’d probably still attempt to get all connectivity and support designed into a spatial visualisation tool like PNI rather than CAD. Afterall, connectivity of any type can be modelled as nodes and arcs (same as PNI). It’s just that ducting tends to have a greater 3D heft than a single line / arc of a typical comms cable. 

Why is it important to have this data in a single spatial system rather than CAD? Well, I figure it should help future augmented reality (AR) use-cases like the ones described in the link.

So here’s the updated diagram:

* There are of course multi-site DC organisations that have links between their sites, but they tend to outsource their long-haul network links to traditional carriers.

The common data store trend

Some time back, we discussed  A modern twist on OSS architecture that is underpinned by a common data model.
 
Time to discuss this a little more visually.
 
As the blue boxes on the left side of the diagram below show, you may have many different data sources (some master, some slaved). You may have a single OSS tool (monolithic solution) or you may have many OSS tools (best-of-breed approach).
 
You may have multiple BSS, NMS and even direct connections to network devices. You may even have other sources of data that you’ve never used before such as weather patterns, lightning strikes, asset management prediction modelling, SCADA data, HVAC data, building access / security events, etc, etc.
 
The common data model allows you to aggregate those sets to provide insights that have never been readily accessible to you previously.
 
So let’s look at a few key points
  1. Existing network layer systems (eg NMS, NE and their mediation devices) are currently sucking (near)real-time (ie alarm and perf) data out of the network and feeding to an OSS directly. They may also be pushing inventory discovery data to the OSS, although probably only loading less frequently (once-daily typically) .
  2. The common data model provides a few options for data flows: 
    1. If the data store is performant enough, the network layer could feed real-time data to the data store which on-forwards to OSS
    2. multi-home the data from the network to the data store and OSS simultaneously
    3. feed data from the network to the OSS, which may (or not) process before pushing to the data store
  3. Just a quick note regarding data flows: The network will tend to be the master for real-time / assurance flows. However, manual input tends to be the master for design/fulfil flows, so the OSS becomes the master of inventory data as per this link 
  4. The question then becomes where the data enrichment happens (ie appending inventory-related data to alarms) to help with root-cause and service-impact calculations. Enrichment / correlation probably needs to happen in the OSS‘s real-time engine, but it could source enrichment data directly from the network, from the OSS‘s inventory, or from the common data store 
  5. If the modern ETL tools (eg SNMP and syslog collectors, etc) allow you to do your own ETL to a common data store, a vendor OSS would only need one mediation device (ie to take data from the data store), rather than needing separate ones to pull from all the different NMS/EMS/NE) in your network. This has the potential to reduce mediation license costs from your OSS vendor
  6. Having said that, if you have difficult / proprietary interfaces that make it a challenge to do all of your own ETL then it might be best to let your OSS vendor build your mediation / ETL engines
  7. The big benefit of the common data store is you can choose a best-of-breed approach but still have a common data model to build Business Intelligence queries and reports around
  8. The common data store also takes load off the production OSS application / data servers. Queries and reports can be run against the common data platform, freeing up CPU cycles on the OSS for faster user interactions

The Common Data Model is supported by a few key advancements:

  1. In the past, the mediation layer (ie getting data out of the network and into the OSS) was a challenge. Network operators didn’t tend to want to do this themselves. This introduced a dependency on software suppliers / integrators to build mediation devices and sell them to operators as part of their OSS/BSS solutions. But there’s been a proliferation of highly scalable ETL (Extract, Transform, Load) tools in recent years
  2. Many networks used to have proprietary interfaces that required significant expertise to integrate with. The increasing ubiquity of IP networking and common interfaces (eg SNMP and web interfaces like RESTful, JSON, SOAP, XML) to the network layer makes ETL simpler.=
  3. Massively scalable databases that don’t have as much dependency on relational integrity and can ingest data for myriad sources
  4. A proliferation of data visualisation tools that are more user-friendly instead of having to be a coder capable of writing complex SQL queries
 

An Asset Management / Inventory trick

Last week we discussed the nuances between Inventory, Asset and Config Management within an OSS stack. Each one of these tools are designed to supports functionality for different users / persona-groups. However, they also tend to have significant functional overlap. Chances are your organisation doesn’t have separate dedicated tools for each.

So today I’m going to share a trick I’ve used in the past when I’ve only had a PNI (Physical Network Inventory) system to work with, but need to perform asset management style functionality.

Most inventory tools are great at storing the current state of a device that exists in a network. However, they don’t tend to be so great at an asset manager’s primary function – tracking the entire life-cycle of an asset from procurement to decommissioning and sparing / maintenance along the way.

Normally the PNI just records the locations of all the active network equipment – in buildings, exchanges, comms-huts, cabinets, etc. The trick I use is to create an additional location/s for warehouses. They may (or may not) reside in the physical location of your real warehouse/s.

In almost all PNI systems, you have control over the status of the device (eg IN-SERVICE, etc). You can use this functionality to include status of SPARE, UNDER REPAIR, etc and switch a device between active network locations and the warehouse.

These status-change records give you the ability to pin-point the location of a given asset at any point in time. It also gives you trending stats, either as an individual device or as a cohort of devices (eg by make/model).

You can even build processes around it for check-in / check-out of the warehouse and maintenance scheduling.

I should point out that this works if your PNI allows you to uniquely identify a device (eg by make/model + serial number or perhaps a unique naming convention instance). If your PNI device records only show the current function of a device (eg a naming convention like SiteA-Router-0001), then you might lose sight of the device’s trail when it moves through life-cycle states (eg to the warehouse).

The differences between Inventory, Asset and Config Management in an OSS

We recently discussed the differences between PNI (Physical Network Inventory) and LNI (Logical Network Inventory) solutions that appear as part of many OSS stacks. 

As promised, today we’ll talk about the subtle differences between:

  • Inventory Management Systems 
  • Asset Management Systems and
  • Configuration Management Databases (CMDB)
  • We might even discuss Virtual Infrastructure (VIM) and Resource Managers as well as Config Managers (different from CMDB) too

Inventory vs Asset vs CMDB

To be honest, the diagram above doesn’t show adequate overlap. Each of these systems has a slightly different purpose, usually for a slightly different set of personas. However, they all play a part in managing the resources that make up an organisation’s Active Network (the network segment dedicated to carrying customer traffic, as opposed to internal corporate traffic).

Let’s start with Inventory Management Systems (IMS) because IMHO, these are the tools that were traditionally responsible for managing service-provider networks. These are the tools typically used by network planners, network engineers, capacity planners and other back-office operational staff.  As mentioned in the link above, these tools can be further broken down into:

  • PNI (Physical Network Inventory) – The physical devices like switches, routers, firewalls as well as the outside plant (OSP) like cables, joints, etc. Generally only used by operators with large, wide-spread networks of physical assets, especially outside plant.
  • LNI (Logical Network Inventory) – The set of objects that are formed using physical infrastructure (and possibly associations to other logical objects). This could include circuits, VLANs, and other overlay network topologies as well as the management of attributes like bandwidth, protocols and other network functionality

These tools tend to focus on the key physical/logical/virtual resources that comprise an operator’s active network (AN). However, they often also support functionality that crosses into other domains such as asset and config management.

Asset Management Systems (AMS), as the name implies, have a more “financial” purpose; where assets are objects of intrinsic financial value to an organisation. AMS tools tend to be used by the accounting and asset management teams. They’re used to track current value (purchase price minus depreciation), warranties, spares management, life-cycles / refresh / end-of-life of assets and their contracts, as well as reactive and predictive maintenance. AMS will tend to store information about most of the Active Network Physical devices. This means they will have records for the same devices as PNI, but often with different information / attributes. They won’t tend to store LNI-related data. However, AMS will often keep information about assets in addition to Active Network devices. This could include software licenses and more.

Configuration Management Databases (CMDB) is more of an IT Service Management (ITSM) terminology. Like many IT concepts, ITSM has been increasingly used in parts of service provider networks. CMDBs are a database of Configuration Items (CIs), where CIs can be logical or physical entities. CIs may (or may not) be physical devices (PNI) or logical resource entities (LNI) and may (or may not) represent tangible values (assets). The main purpose of CIs is to store information about IT services that will allow other ITSM processes, such as Incident, Problem and Change Management, to be performed efficiently.

Not only is there functional overlap between these systems, there’s often also terminology overlap and/or misalignments. Different vendors have different levels of functionality and support alternate use-cases, so the areas of overlap differ between organisations.

Oh, and I also promised to mention VIMs and Config Managers:

Virtual Infrastructure Managers (VIM) are responsible for managing the virtual resources made available by physical infrastructure like compute, storage and network devices. In some cases, VIMs generate virtual network devices (VNFs) or virtual machines (VMs) that could look almost identical to any other device stored in LNI, PNI, AMS and/or CMDB. In fact, instances of these VNFs and VMs may even appear in those systems.

Config Management (as opposed to, but also potentially overlapping with, CMDB), is all about managing the configurations of devices in the network (often active network and corporate network). Each device, such as a router, has a configuration that tells the hardware how to function, where to route traffic, which packets to prioritise, where to send management logs (to the OSS), etc. Being able to monitor and manage these configurations centrally and consistently is the purpose for Config Managers. These are mostly used by network engineers to set policies and golden-configs (ie the config templates that all devices of that type must adhere to consistently). For example, you may have hundreds/thousands of devices in your network and want to re-point all management traffic to a new server as part of an OSS upgrade. Rather than configuring each device separately and manually, you can use the config management tool to push config changes out to the network.

Leave us a message to describe how your organisation use these (and other) tools.

OSS discovers a network

Following yesterday’s post about OSS Inventory, I received another great follow-up question from another avid reader of the PAOSS blog:

Interesting thoughts Ryan! In addition to ‘faults up’, perhaps there is a case also (obvious?) for ‘discovery up’ to capture ongoing non-planned changes? Wondering have you come across any sort of reconciliation / adaptive inventory patterns like this? Workflow based? Autonomous? (Going to far into chaos theory territory ?

Yes, we did exactly that with the same tool discussed yesterday that I used back in 2000. In fact, a very clever dev and I got that company’s first-ever auto-discovery tool working on site (using a product supplied by head-office). Discovering the nodals (ie equipment, cards, ports) was fairly easy. Discovering the connectivity within a domain (we started with SDH) was tricky, but achievable. Auto-discovering cross-domain connectivity (ie DSL circuits through physical, SDH transit, ATM and logical connectivity onto the IP cloud) was much trickier as we needed to find/make linking keys across different data sources.

It was definitely workflow based with a routine-driven back-end. We didn’t just want anything that was discovered to be automatically stuffed into (or removed from) the database (think flapping ports or equipment going down temporarily). It could’ve been automous, but we introduced a manual step to approve any of the discoveries made by each automated discovery iteration.

As you know, modern networks / EMS / VIM (resource managers) are much more discoverable. They need to be for modern orchestration and resilience techniques. I don’t think it would be quite so tricky to stitch circuits together as we’re no longer so circuit-oriented as back in 2000.

However, I’d be fascinated to hear from other readers how much of a problem they have trying to marry up different data sources for discovery purposes. I’d also love to hear whether they’re using fully autonomous discovery or using the manual intervention step for the same reason we were. I imagine most are automating, because orchestration plans just need to make use of whatever resources are being presented by the underlying resource managers in near-real-time.

PS. For those wondering what “discovery” is, it’s shown in the lower grey arrow in this diagram from “Orders down, Faults up

Discovery is the process that allows data to be passed from NMS/EMS/NEs (ie the network or resource managers) directly into the inventory management database. It should be a more reliable and expedient way of sychronising the inventory with the live network. 

The reason for the upper grey arrow is because not all networks have APIs that can be “discovered.” Passive equipment like cable joints and patch-panels don’t have programmatic interfaces. Therefore we need to find other ways to get that data into the Inventory Manager.

Various forms of OSS Inventory

After reading other recent posts such as “Orders Down, Faults Up” and “How is OSS/BSS service and resource availability supposed to work?” an avid reader of the PAOSS blog posed the following brilliant question:

Do you have any thoughts on geospatial vs non geospatial network inventory systems? How often do you see physical plant mapping in a separate system from network inventory, with linkages or integrations between them, vs how often do you see physical and logical inventory being captured primarily in a geospatially oriented system?

Boy do I ever have some thoughts on this topic!! I’m sure you do too, so I’d love to hear what you think in the comments section below.

I was lucky. The first OSS/BSS that I worked on (all the way back in 2000), had both geo and non-geo (topology) views. It also had a brilliantly flexible data model that accommodated physical and logical inventory. All tightly integrated into one package. There aren’t many tools that can do all of that even today. Like I said, I was lucky to have this as a starting point!!

Like all things OSS/BSS, it starts with the personas and the key tasks they need to perform. Or from the supplier’s perspective, which customer personas they’re most actively targeting.

For example, if you have a significant Outside Plant (OSP) Network, then geo-positioning is vital. The exchanges and comms huts are easy enough to find, but pits, cable routes, easements, etc are often harder to find. It’s not uncommon for a field tech to waste time searching for a pit that’s covered in dirt, grass or snow. And knowing the exact cable route in geo view is helpful for sending field techs to the exact location of a fault (ie helping them to pinpoint the location of the bright yellow excavator that has just sliced through your inter-capital link). Geo-view is also important for OSP designers and the field workforce that builds the OSP network.

But other personas don’t care about seeing the detailed cable route. They just want to see a point-to-point topological link to represent physical connections between the ports on adjacent devices. This helps them to quickly understand the network or circuit / service view. They may also like to see an alarm overlay on the topology to quickly determine which parts of the network aren’t performing as expected. For these personas, seeing all the geo-detail just acts as visual noise that they need to subconsciously filter out to understand the topology view.

These personas also tend to want topological views of the network, not just the physical but the logical and virtual network / service overlays too.

In most cases that I can think of, the physical / OSP inventory tools show the physical devices (ports even) that the OSP network connects into. Their main focus is on the cables, joints, pits, pipes, catenaries, poles, lead-ins, patch-panels, patch-leads, splitters, etc. But showing the termination of cables onto active equipment (Inside Plant or ISP) is an important linking key between the physical and logical views.

The physical port (on the physical device) becomes the key demarcation between physical and logical worlds. The physical port connects physical cables / leads, but it also acts as the anchor point from which to create logical ports to which logical connections are made. As a result, the physical device and port tend to be shown in both physical (geo) and logical inventory tools. They also tend to be shown in both physical and logical network topology views.

In the case of the original OSS/BSS I worked on, it had separate visualisation tools for geo, network and circuit/service, but all underpinned by a common data model.

What’s the best way? Different personas will have different perspectives of course. I prefer for physical and logical inventories to be integrated out of the box (to allow simple cross-ref visually and in queries)…. but I also prefer for them to have different views (eg geo, topology, network, circuit/service) to suit different situations.

I also find it helpful if each of those views allow the ability to drill down deeper into specific sections of the graph if necessary. I’d prefer not to have all of those different views overlaid onto a geo visualisation. Too much visual clutter IMHO, but others may love it that way.

Oh, and having separate LNI (Logical Network Inventory) and PNI (Physical Network Inventory) can be a tricky thing to reconcile. The LNI will almost always have programmatic interfaces (APIs) to collect data from, but will generally have to amalgamate many different sources. Meanwhile, the PNI consists of mostly passive equipment and therefore has no API to collect latest info from. I tend to use strategies at the above-mentioned demarcation point (ie physical ports) to help establish linking keys between LNI and PNI.

BTW. There’s one aspect of the question, “How often do you see physical plant mapping in a separate system from network inventory” that I haven’t fully answered. I’ll cover the question of asset management vs inventory management vs CMDB (Configuration Management Database) in more detail in an upcoming post. [Ed. See link here]

In need of an OSS transformation translator

As OSS Architects, we have an array of elegant frameworks to call upon when designing our transformational journeys – from current state to a target state architecture.

For example, when providing data mapping, we have tools to prepare current and/or target-state data diagrams such as the following:

Source here.

These diagrams are really elegant and powerful for communicating with other data experts and delivery teams. It’s a data expert language.

Data experts are experts of the ETL (Extract, Transform, Load) process, but often have less expertise with the actual meaning and importance of the data sets themselves. For example, a data expert may know there’s a product offerings table, and each has 23 associated attributes (eg bandwidth, SLA class, etc) available. But they may have less understanding of the 245 product types that are housed in the product’s data table, and even less awareness of the meanings of the thousands of product attributes. You need to be a subject matter expert (SME) to understand that detail about the data. In some cases, the SME might be from your client and knows far more tribal knowledge than you.

We often need other SMEs (the products expert in this case) to help us understand what has to happen with the data during transformation. What do we keep, what do we change, what do we discard, etc.

Just one problem – SMEs might not always speak the same language as the data experts.

As elegant as it is, the data relationships diagram above might not be the most intuitive format for product experts to review and comment.

As with many aspects of Architecture and transformation, if we’re to understand, it’s best to communicate in our audience’s language.

In this case, it might be best to show data mappings as overlays on screenshots that the Product owner is familiar with:

  • From
    • Their current GUI
    • Existing sales order forms
    • Current report templates
  • To
    • Their next-generation GUI
    • New order forms
    • Post-Transform report templates

Such an approach might not look elegant to our data expert colleagues. The question is whether it quickly makes enough sense to the SMEs for you to elicit concise responses from them.

The “right” approach is not always the most effective.

I’d love to hear your tips, tricks and recommendations for speaking / listening in the audience’s language.

Orders down, faults up

As mentioned in a post about Service and Resource Availability last week, I do tend to think of OSS workflows around an “orders down, faults up,” flow direction. And that means customers (services) at the top, network (resources) at the bottom of the (TMN) pyramid.

I also think of inventory (yellow) as the point where Assurance / Faults (blue) and Fulfillment / Orders (purple) collide and enhance, as per the diagram below.

These are highly generic examples, but let’s take a closer look:

Assurance flow (blue) – an alarm or event in the network (NEL/NE layer) pushes up through the stack to the OSS as a fault. The inventory (network / service) helps to enrich the fault with additional information (eg device name, location, connectivity, correlation of this and other faults, etc) to help resolve the fault (either manually by operators or automatically by algorithms). It also helps associate the fault in the network/resource with the customer/s using those resources. This allows notifications to be issued to customers. Note that this simple flow doesn’t reflect examples such as an incident (ie when a customer notices a problem first and calls it in before the OSS has been able to issue a notification).

Fulfillment flow (purple) – a customer places an order (BML/BSS layer or above) and it pushes down through the stack, including changes in the network (NEL/NE layer). Once all the appropriate network changes have been made, the order is ready for use by the customer. Once again, inventory plays an important part, associating customer / service identifiers with suitable resources from the available resource pool. Generally the (customer facing) service orders won’t have the technology-specific details required to actually update the network configurations though. That’s where the inventory often helps to fill in the knowledge gaps and send technology-specific commands down into the network. [See Friday’s post for more information about CFS and RFS definitions and mappings]

Inventory flows (yellow) – an inventory is relevant to assurance and fulfillment flows if the BSS and network / resource layers don’t hold enough information to be fully processed. The enriching information stored by inventory must come from somewhere. Some of it comes from Discovery (usually an automated process of collecting from the network or other sources), or via Manual / Scripted Input (eg physical network designs including patch cables and splicing). Some data (eg splices) just can’t be collected automatically as they relate to passive equipment that has no programmatic interface. This data just has to be created manually.
But arguably the more important inventory data is actually recording the mappings made from customers (services) to network (resources). Inventory solutions  are often where these linking keys / relationships are recorded.

These flows also tend to indicate the direction of data mastery. Whilst the network itself is the source of truth, Fulfillment flows will start at the BSS and customer / service / order data will tend to be mastered there before orders are pushed into the network as provisioning commands. For assurance flows, the network will tend to be the master source of data, but with enrichment along it’s path northbound.

Just keep in mind that there are many exceptions to these examples. Data and processes can flow in many different ways. The diagram above is just useful for helping newcomers to understand some conceptual processes and data models / flows.

Differences between CFS and RFS

Further to yesterday’s post about Service and Resource Availability, I received some questions about how to discern between CFS (Customer Facing Services) and RFS (Resource Facing Services).

I thought the following description, sourced from TM Forum’s GB999 (Service overview sec 1.1.3), might help clarify the differences:

  • “This enables us to model a wide variety of services in a common class hierarchy while differentiating between Services that are obtained as a Product by a Customer versus those that aren’t. As we will see, a CustomerFacingService is one that is obtained as a Product by a Customer. Therefore, the Customer may have specific control over this Service via its associated Product. In contrast, the Customer never knows explicitly which ResourceFacingServices are being used to support a CustomerFacingService. More importantly, the Customer shouldn’t have to know which ResourceFacingServices are being used, since the Customer hasn’t explicitly obtained them.”
  • CFS are associated with resource technology neutral services i.e. they describe general capabilities and have attributes that are general across many technologies e.g. throughput, latency, SLA /loss rate, availability.
  • CFS and RFS typically have different lifecycles, CFS are related to customer and product changes and RFS to technology changes.
  • RFS are associated with resource technology specific services i.e. they have attributes that predominately relate to a specific technology.
  • RFS typically do some of the following:
    1. Map between the native protocols used to expose management of resources e.g. Netconf, YANG, SNMP, etc.
    2. Provide some integrated approach to provisioning and assuring RFS that span multiple technical domains e.g. slices across RAN and Core).
    3. They may be Operator, SP, ISV or Supplier provided.

Further important notes:

  • Composition of subordinate CFSs to support the CFSs exposed by Production capabilities (iterative composition pattern). These subordinate CFSs may be from other Operational Domains both within the same operator or acquired from third party operators as happens with wholesale interconnect.
  • Mapping of CFS to internal Resource Facing Services (RFS) that abstract into services the resource defined by Suppliers’ Technical Domains whose boundaries are defined by technology and supplier choices. This mapping links the boundary decisions of Operational Domains to the Technical Domain boundary decisions of suppliers.
  • RFSs can be atomic or composite to include other RFS (iterative composition pattern). This is a decision taken by the Operations / Integrator composing or creating RFSs based on deployment needs.
  • In a Service Oriented Architecture any exposed services can be consumed by any other service.

How is OSS/BSS service and resource availability supposed to work?

The brilliant question above was asked by a contact who is about to embark on a large OSS/BSS transformation.  That’s certainly a challenging question to start the new year with!!

The following was provided for a little more context:

  • We have a manually maintained table for each address where we can store which services are available—ie. DSL up to 5 Mbps or Fiber Data 300 Mbps

  • This manual information has no data-level connection to the actual plant used to serve the address 

  • In a “perfect world”, how does this work?

  • Where is the data stored? Ex: Does a geospatial network inventory store this data, then the BSS requests it as needed?

  • How does a typical OSS tie together physical network and equipment to products and offerings?

  • How is it typically stored? How is it accessed?

  • Sort of related to the address, we have “Facility” records that include things like the Electronics (Card, slot, port, shelf, etc) and some important “hops” along the way 

  • Right now if a tech makes changes to physical plant, we have to manually update our mapping (if the path changes), spreadsheets (if fiber assignment changes) or paper records (if copper pair assignments change).. additionally, we might need to update the Facilities database

  • It doesn’t “use” it’s “awareness” of our plant or network equipment to do anything except during service orders where certain products are tied to provisioning features—ie. callerID billing code being on an order causes a command to be issued to the switch to add that feature.

  • There is no visibility into network status.. how does this normally work?

  • I feel like I’m missing a fundamental reference point because I’ve never seen an actual working example of “Orders down, faults up”, just manually maintained records that sort of single-directionally “integrate” to network devices but only in the context of what was ordered, not in the context of what is available and what the real-time status is.

Wow! Where do we start? Certainly not an easy or obvious question by any means. In fact it’s one of the trickier of all OSS/BSS challenges.

In the old days of OSS/BSS, services tended to be circuit-oriented and there was a direct allocation of logical / physical resources to each customer service. You received an order, you created a “customer circuit” for the order, you reserved suitable / available resources in your inventory to assign to the circuit, then issued work order activities to implement the circuit. When the work order activities were complete, the circuit was ready for service.

The utilised resources in your inventory system/s were tagged with the circuit ID or service ID and therefore not available to other services. This association also allowed Service Impact Analysis (SIA) to be performed. In the background, you had to reconcile the real resources available in the network with what was being shown in your inventory solution. Relationships were traceable down through all layers of the TMN stack (as below). Status of the resources (eg a Network Element had failed) could also be associated to the inventory solution because alarms / events had linking keys to all the devices, cards, ports, logical ports, etc in inventory .

To an extent, it’s still possible to do this for the access/edge of the network. For example, from the Customer Premises Equipment (CPE) / Network Termination Device (NTD) to the telco’s access device (eg DSLAM or similar). But from that point deeper into the core of the telco network, it’s usually a dynamic allocation of resources (eg packet-switched, routed signal paths).

With modern virtualised and packet-switched networks, dynamic allocation makes its harder to directly associate actual resources with customer services at any point in time. See this earlier post for more detail on the diagram below.

Instead, we now just ask the OSS to push orders into the virtualisation cloud and expect the virtualisation managers to ensure reliable availability of resources. We’ve lost visibility inside the cloud.

So this poses the question about whether we even need visibility now. There are three main states to consider:

  1. At Service Initiation – What resources are available to assign to a service? As long as capacity planning is doing its job and keeping the available resource pool full, we just assume there will be sufficient resource and let the virtualisation manager do its thing
  2. After Service is Commissioned – What resources are assigned to the service at the current point in time? If the virtualisation manager and network are doing their highly available, highly resilient thing, then do we want to know?
  3. During an Outage – What services are impacted by resources that are degraded or not available? As operators, we definitely want to know what needs to be fixed and which customers need to be alerted.

So, let’s now get into a more “modern orchestration and abstraction” approach to associating customer services with resources. I’ve seen it done many different ways but let’s use the diagram below as a reference point (you might have to view in expanded form):

CFS RFS orchestration

 Here are a few thoughts that might help:

  • As mentioned by the contact, “orders down, faults up,” is a mindset I tend to start with too. Unfortunately, data flows often have to be custom-designed as they’re constrained by the available systems, organisation structures, data quality improvement models, preferred orchestration model, etc
  • You may have heard of CFS (Customer Facing Service) and RFS (Resource Facing Service) constructs? They’re building blocks that are often used by operators to design product offerings for customers (and then design the IT systems that support them). They’re shown as blue ovals as they’re defined in the Service Catalog (CFS shown as north-facing and RFS as south-facing)
  • CFS are services tied to a product/offering. RFS are services linked with resources
  • To simplify, I think of CFS like a customer order form (ie what fields and options are available for the customer) and RFS being the technical interfaces to the network (eg APIs into the Domain Managers and possibly NMS/EMS/VIM)
  • Examples of CFS might be VPN, Internet Access, Transport, Security, Mobility, Video, etc
    Examples of RFS might be DSL, DOCSIS, BGP (border gateway protocol), DNS, etc
    See conceptual model from Oracle here:
  • Now, let’s think of how to create this model in two halves:
      • One is design-time – that’s where you design the CFS and/or RFS service definitions, as well as designing the orchestration plan (OP) (aka provisioning plan). The OP is the workflow of activities required to activate a CFS type. This could be as simple as one CFS consuming an RFS stub with a few basic parameters mapped (eg CallingID). Others can be very complex flows if there are multiple service variants and additional data that needs to be gathered from East-West systems (eg request for next available patch-port from physical network inventory [PNI]). Some of the orchestration steps might be automated / system-driven, whilst others might be manual work order activities that need to be done by field workforce.
        Note that the “Logging and Test” box at the left is just to test your design-time configurations prior to processing a full run-time order
      • The other is run-time – that’s where the Orchestrator runs the OP to drive instance-by-instance implementation of a service (including consumption of actual resources). That is, an instantiation of one customer order through the orchestration workflow you created during design-time 
  • A CFS can map parameters from one or more RFS (there can even be hierarchical consumption of multiple RFS and CFS in some situations, but that will just confuse the situation)
  • You can also loosely think of CFS as being part of the BSS and RFS as being part of the OSS, with the service orchestration usually being a grey area in the middle
  • Now to the question about where is the data stored:
    • Design-time – CFS building block constructs are generally stored in a BSS or service catalog. Orchestration plans are often also part of modern catalogs, but could also fall within your BSS or OSS depending on your specific stack
    • Run-time (ie for each individual order) – The customer order details (eg speeds, configurations, etc) are generally stored in “the BSS.” The orchestration plan for each order then drives data flows. This is where things get very specific to individual stacks. The OP can request resource availability via east-west systems (eg inventory [LNI or PNI], DNS, address databases, WFM, billing code database, etc, etc, etc) and/or to southbound interfaces (eg NMS/EMS/Infrastructure-Manager APIs) to gather whatever information is required
    • Distributed or Centralised data – There’s no specific place where all data is collected. Some of the systems (eg PNI/LNI) above will have their own data repositories, whilst others will pull from a centralised data store or the network or other infrastructure via NMS/EMS/VIM
    • Data master – in theory the network (eg NMS/EMS/NE) should be the most accurate store of information, hence the best place to get data from (and your best visibility of current state of the network). Unfortunately, the NMS/EMS/NE often won’t have all the info you need to drive the orchestration plan. For example, if you don’t already have a cable to the requesting customer’s address, then the orchestration plan will have to include an action/s for a designer to use PNI/geospatial data to find the nearest infrastructure (eg joint/pedestal) to run a new cable from, then go through all the physical build activities, before sending the required data back to the orchestration plan. Since the physical network (eg cables, joints, etc) almost never has a programmatic interface, it will require manual effort and manual data entry. Alternatively, the NMS/EMS/VIM might not be able to tell us exactly what resource the service is consuming at any point in time
    • For Specific product offerings – There are so many different possibilities here that it’s hard to answer all the possible data flows / models. The orchestration plan within the Business Orchestration (aka Cross-domain Orchestration) layer is responsible for driving flows. It may have to perform service provisioning, network orchestration and infrastructure control. 

This is far less concise than I hoped. 

If you have a simpler way of answering the question and/or can point us to a better description, we’d love to hear from you!

What’s in your OSS for me?

May I ask you a question?  Do the senior executives at your organisation ever USE your OSS/BSS?

I’d love to hear your answer.

My guess is that few, if any, do. Not directly anyway. They may depend on reports whose data comes from our OSS, but is that all?

Execs are ultimately responsible for signing off large budget allocations (in CAPEX and OPEX) for our OSS. But if they don’t see any tangible benefits, do the execs just see OSS as cost centres? And cost centres tend to become targets for cost reduction right?

Building on last week’s OSS Scoreboard Analogy, the senior execs are the head coaches of the team. They don’t need the transactional data our OSS are brilliant at collating (eg every network device’s health metrics). They need insights at a corporate objective level.

How can we increase the executives’s, “what’s in it for me?” ranking of the OSS/BSS we implement? We can start by considering OSS design through the lens of senior executive responsibilities:

  • Strategy / objective development
  • Strategy execution (planning and ongoing management to targets)
  • Clear communication of priorities and goals
  • Optimising productivity
  • Risk management / mitigation
  • Optimising capital allocation
  • Team development

And they are busy, so they need concise, actionable information.

Do we deliver functionality that helps with any of those responsibilities? Rarely!

Could we? Definitely!

Should we? Again, I’d love to hear your thoughts!

 

The Ineffective OSS Scoreboard Analogy

Imagine for a moment that you’re the coach of a sporting team. You train your team and provide them with a strategy for the game. You send them out onto the court and let them play.

The scoreboard gives you all of the stats about each player. Their points, blocks, tackles, heart-rate, distance covered, errors, etc. But it doesn’t show the total score for each team or the time remaining in the game. 

That’s exactly what most OSS reports and dashboards are like! You receive all of the transactional data (eg alarms, truck-rolls, device performance metrics, etc), but not how you’re collectively tracking towards team objectives (eg growth targets, risk reduction, etc). 

Yes, you could infer whether the team is doing well by reverse engineering the transactional data. Yes, you could then apply strategies against those inferences in the hope that it has a positive impact. But that’s a whole lot of messing around in the chaos of the coach’s box with the scores close (you assume) and the game nearing the end (possibly). You don’t really know when the optimal time is to switch your best players back into the game.

As coach with funding available, would you be asking your support team to give you more transactional tools / data or the objective-based insights?

Does this analogy help articulate the message from the previous two posts (Wed and Thurs)?

PS. What if you wanted to build a coach-bot to replace yourself in the near future? Are you going to build automations that close the feedback loop against transactional data or are you going to be providing feedback that pulls many levers to optimise team objectives?

One big requirement category most OSS can’t meet

We talked yesterday about a range of OSS products that are more outcome-driven than our typically transactional OSS tools. There’s not many of them around at this stage. I refer to them as “data bridge” products.
 
Our typical OSS tools help manage transactions (alarms, activate customers services, etc). They’re generally not so great at (directly) managing objectives such as:
  • Sign up an extra 50,000 customers along the new Southern network corridor this month
  • Optimise allocation of our $10M capital budget to improve average attainable speeds by 20% this financial year
  • Achieve 5% revenue growth in Q3
  • Reduce truck rolls by 10% in the next 6 months
  • Optimal management of the many factors that contribute to churn, thus reducing churn risk by 7% by next March
 
We provide tools to activate the extra 50,000 customers. We also provide reports / dashboards that visualise the numbers of activations. But we don’t tend to include the tools to manage ongoing modelling and option analysis to meet key objectives. Objectives that are generally quantitative and tied to time, cost, etc and possibly locations/regions. 
 
These objectives are often really difficult to model and have multiple inputs. Managing to them requires data that’s changing on a daily basis (or potentially even more often – think of how a single missed truck-roll ripples out through re-calculation of optimal workforce allocation).
 
That requires:
  • Access to data feeds from multiple sources (eg existing OSS, BSS and other sources like data lakes)
  • Near real-time data sets (or at least streaming or regularly updating data feeds)
  • An ability to quickly prepare and compare options (data modelling, possibly using machine-based learning algorithms)
  • Advanced visualisations (by geography, time, budget drawdown and any graph types you can think of)
  • Flexibility in what can be visualised and how it’s presented
  • Methods for delivering closed-loop feedback to optimise towards the objectives (eg RPA)
  • Potentially manage many different transaction-based levers (eg parallel project activities, field workforce allocations, etc) that contribute to rolled-up objectives / targets
 
You can see why I refer to this as a data bridge product right? I figure that it sits above all other data sources and provides the management bridge across them all. 
 
PS. If you want to know the name of the existing products that fit into the “data bridge” category, please leave us a message.

Do you want funding on an OSS project?

OSS tend to be very technical and transactional in nature. For example, a critical alarm happens, so we have to coordinate remedial actions as soon as possible. Or, a new customer has requested service so we have to coordinate the workforce to implement certain tasks in the physical and logical/virtual world. When you spend so much of your time solving transactional / tactical problems, you tend to think in a transactional / tactical way.
 
You can even see that in OSS product designs. They’ve been designed for personas who solve transactional problems (eg alarms, activations, etc). That’s important. It’s the coal-face that gets stuff done.
 
But who funds OSS projects? Are their personas thinking at a tactical level? Perhaps, but I suspect not on a full-time basis. Their thoughts might dive to a tactical level when there are outages or poor performance, but they’ll tend to be thinking more about strategy, risk mitigation and efficiency if/when they can get out of the tactical distractions.
 
Do our OSS meet project sponsor needs? Do our OSS provide functionality that help manage strategy, risk and efficiency? Well, our OSS can help with reports and dashboards that help them. But do reports and dashboards inspire them enough to invest millions? Could sponsors rightly ask, “I’m spending money, but what’s in it for me?”
 
What if we tasked our product teams to think in terms of business objectives instead of transactions? The objectives may include rolled-up transaction-based data and other metrics of course. But traditional metrics and activities are just a means to an end.
 
You’re probably thinking that there’s no way you can retrofit “objective design” into products that were designed years ago with transactions in mind. You’d be completely correct in most cases. So what’s the solution if you don’t have retrofit control over your products?
 
Well, there’s a class of OSS products that I refer to as being “the data bridge.” I’ll dive into more detail on these currently rare products tomorrow.

An OSS checksum

Yesterday’s post discussed two waves of decisions stemming from our increasing obsession with data collection.

“…the first wave had [arisen] because we’d almost all prefer to make data-driven decisions (ie decisions based on “proof”) rather than “gut-feel” decisions.

We’re increasingly seeing a second wave come through – to use data not just to identify trends and guide our decisions, but to drive automated actions.”

Unfortunately, the second wave has an even greater need for data correctness / quality than we’ve experienced before.

The first wave allowed for human intervention after the collection of data. That meant human logic could be applied to any unexpected anomalies that appeared.

With the second wave, we don’t have that luxury. It’s all processed by the automation. Even learning algorithms struggle with “dirty data.” Therefore, the data needs to be perfect and the automation’s algorithm needs to flawlessly cope with all expected and unexpected data sets.

Our OSS have always had a dependence on data quality so we’ve responded with sophisticated ways of reconciling and maintaining data. But the human logic buffer afforded a “less than perfect” starting point, as long as we sought to get ever-closer to the “perfection” asymptote.

Does wave 2 require us to solve the problem from a fundamentally different starting point? We have to assume perfection akin to a checksum of correctness.

Perfection isn’t something I’m very qualified at, so I’m open to hearing your ideas. 😉

 

Riffing with your OSS

Data collection and data science is becoming big business. Not just in telco – our OSS have always been one of the biggest data gatherers around – but across all sectors that are increasingly digitising (should I just say, “all sectors” because they’re all digitising?).

Why do you think we’re so keen to collect so much data?

I’m assuming that the first wave had mainly been because we’d almost all prefer to make data-driven decisions (ie decisions based on “proof”) rather than “gut-feel” decisions.

We’re increasingly seeing a second wave come through – to use data not just to identify trends and guide our decisions, but to drive automated actions.

I wonder whether this has the potential to buffer us from making key insights / observations about the business, especially senior leaders who don’t have the time to “science” their data? Have teams already cleansed, manipulated, aggregated and presented data, thus stripping out all the nuances before senior leaders ever even see your data?

I regretfully don’t get to “play” with data as much as I used to. I say regretfully because looking at raw data sets often gives you the opportunity to identify trends, outliers, anomalies and patterns that might otherwise remain hidden. Raw data also gives you the opportunity to riff off it – to observe and then ask different questions of the data.

How about you? Do you still get the opportunity to observe and hypothesise using raw OSS/BSS data? Or do you make your decisions using data that’s already been sanitised (eg executive dashboards / reports)?

 

OSS diamonds are forever (part 2)

Wednesday’s post discussed how OPEX is forever, just like the slogan for diamonds.
 
As discussed, some aspects of Operational Expenses are well known when kicking off a new OSS project (eg annual OSS license / support costs). Others can slip through the cracks – what I referred to as OPEX leakage (eg third-party software, ongoing maintenance of software customisations).
 
OPEX leakage might be an unfair phrase. If there’s a clear line of sight from the expenses to a profitable return, then it’s not leakage. If costs (of data, re-work, cloud services, applications, etc) are proliferating with no clear benefit, then the term “leakage” is probably fair.
 
I’ve seen examples of Agile and cloud implementation strategies where leakage has occurred. And even the supposedly “cheap” open-source strategies have led to surprises. OPEX leakage has caused project teams to scramble as their financial year progressed and budgets were unexpectedly being exceeded.
 
Oh, and one other observation to share that you may’ve seen examples of, particularly if you’ve worked on OSS in large organisations – Having OPEX incurred by one business unit but the benefit derived by different business units. This can cause significant problems for the people responsible for divisional budgets, even if it’s good for the business as a whole. 
 
Let me explain by example: An operations delivery team needs extralogging capability so they stand up a new open-source tool. They make customisations so that log data can be collected for all of their network types. All log data is then sent to the organisation’s cloud instance. The operations delivery team now owns lifecycle maintenance costs. However, the cost of cloud (compute and storage) and data lake licensing have now escalated but Operations doesn’t foot that bill. They’ve just handed that “forever” budgetary burden to another business unit.
 
The opposite can also be true. The costs of build and maintain might be borne by IT or ops, but the benefits in revenue or CX (customer experience) are gladly accepted by business-facing units.
 
Both types of project could give significant whole-of-company benefit. But the unit doing the funding will tend to choose projects that are less effective if it means their own business unit will derive benefit (especially if individual’s bonuses are tied to those results).
 
OSS can be powerful tools, giving and receiving benefit from many different business units. However, the more OPEX-centric OSS projects that we see today are introducing new challenges to get funded and then supported across their whole life-cycle.
 
PS. Just like diamonds bought at retail prices, there’s a risk that the financials won’t look so great a year after purchase. If that’s the case, you may have to seek justification on intangible benefits.  😉
 
PS2. Check out Robert’s insightful comment to the initial post, including the following question, “I wonder how many OSS procurements are justified on the basis of reducing the Opex only *of the current OSS*, rather than reducing the cost of achieving what the original OSS was created to do? The former is much easier to procure (but may have less benefit to the business). The latter is harder (more difficult analysis to do and change to manage, but payoff potentially much larger).”

Diamonds are Forever and so is OSS OPEX

Sourced from: www.couponraja.in

I sometimes wonder whether OPEX is underestimated when considering OSS investments, or at least some facets (sorry, awful pun there!) of it.

Cost-out (aka head-count reduction) seems to be the most prominent OSS business case justification lever. So that’s clearly not underestimated. And the move to cloud is also an OPEX play in most cases, so it’s front of mind during the procurement process too. I’m nought for two so far! Hopefully the next examples are a little more persuasive!

Large transformation projects tend to have a focus on the up-front cost of the project, rightly so. There’s also an awareness of ongoing license costs (usually 20-25% of OSS software list price per annum). Less apparent costs can be found in the exclusions / omissions. This is where third-party OPEX costs (eg database licenses, virtualisation, compute / storage, etc) can be (not) found.

That’s why you should definitely consider preparing a TCO (Total Cost of Ownership) model that includes CAPEX and OPEX that’s normalised across all options when making a buying decision.

But the more subtle OPEX leakage occurs through customisation. The more customisation from “off-the-shelf” capability, the greater the variation from baseline, the larger the ongoing costs of maintenance and upgrade. This is not just on proprietary / commercial software, but open-source products as well.

And choosing Agile almost implies ongoing customisation. One of the things about Agile is it keeps adding stuff (apps, data, functions, processes, code, etc) via OPEX. It’s stack-ranked, so it’s always the most important stuff (in theory). But because it’s incremental, it tends to be less closely scrutinised than during a CAPEX / procurement event. Unless carefully monitored, there’s a greater chance for OPEX leakage to occur.

And as we know about OPEX, like diamonds, they’re forever (ie the costs re-appear year after year). 

A billion dollar bid

A few years ago I was lucky enough to be invited to lead a bid. I say lucky because the partner organisations are two of the most iconic firms in the tech industry. The bid was for bleeding-edge work, potentially worth well over a billion dollars. I was a little surprised to be honest. I mean, two tech titans, with many very, very clever people, much cleverer than me. Why would they need to look outside and engage me?

As it turned out, the answer became clear within the first few meetings. And whilst the project had little to do with OSS, it certainly had (has) parallels in the world of OSS.

Both of the organisations were highly siloed. Each product / capability silo had immense talent and immense depth to it. Our combined team had many PhDs who could discuss their own silo for hours, but could only point me in the general direction of what plugged into their products. 

Clearly, I was engaged to figure out the required end-to-end solution for the customer and then how to bolt the two sets of silos into that solution framework.

The same is true when looking for OSS solution gaps, in my experience at least. If you look into a domain or a product, the functionality / capability is usually quite well defined, understood and supported. For example, alarm / event managers are invariably very good at managing alarm / event lists.

If you’re going to find gaps, they’re more likely to be found in the end-to-end solution – in the handoffs, responsibility demarcation points, interfaces and processes that cross between silos. That’s why external consultancies can prove valuable for large organisations. They generally look into the cross-domain solution performance.

As you’d already know, the end-to-end solution is a combination of people, process and technology. Even so, as the “manager of managers,” I’m not sure our OSS tech is solving this problem as well as it could. Is there even a “glue” product that’s missing from our OSS/BSS stack?

Sure, we have some tools that fit this purpose – workflow engines, messaging buses, orchestration engines, data lakes, etc. Yet I still feel there’s an opportunity to do it far better. And the opportunity probably extends far beyond just OSS and into the broader IT industry.

What have you done to help solve this problem on your OSS suites?

PS. If you’re wondering what happened to the bid. Well, the team was excited to have made the shortlist of 3, but then the behemoths decided to withdraw from the race. Turns out that winning the bid could’ve jeopardised the even bigger supply contracts they already had with the client. Boggles the mind to think there were bigger contracts already in play!!

 

Inventory Management re-states its case

In a post last week we posed the question on whether Inventory Management still retains relevance. There are certainly uses cases where it remains unquestionably needed. But perhaps others that are no longer required, a relic of old-school processes and data flows.
 
If you have an extensive OSP (Outside Plant) network, you have almost no option but to store all this passive infrastructure in an Inventory Management solution. You don’t have the option of having an EMS (Element Management System) console / API to tell you the current design/location/status of the network. 
 
In the modern world of ubiquitous connection and overlay / virtual networks, Inventory Management might be less essential than it once was. For service qualification, provisioning and perhaps even capacity planning, everything you need to know is available on demand from the EMS/s. The network is a more correct version of the network inventory than external repository (ie Inventory Management) can hope to be, even if you have great success with synchronisation.
 
But I have a couple of other new-age use-cases to share with you where Inventory Management still retains relevance.
 
One is for connectivity (okay so this isn’t exactly a new-age use-case, but the scenario I’m about to describe is). If we have a modern overlay / virtual network, anything that stays within a domain is likely to be better served by its EMS equivalent. Especially since connectivity is no longer as simple as physical connections or nearest neighbours with advanced routing protocols. But anything that goes cross-domain and/or off-net needs a mechanism to correlate, coordinate and connect. That’s the role the Inventory Manager is able to do (conceptually).
 
The other is for digital twinning. OSS (including Inventory Management) was the “original twin.” It was an offline mimic of the production network. But I cite Inventory Management as having a new-age requirement for the digital twin. I increasingly foresee the need for predictive scenarios to be modelled outside the production network (ie in the twin!). We want to try failure / degradation scenarios. We want to optimise our allocation of capital. We want to simulate and optimise customer experience under different network states and loads. We’re beginning to see the compute power that’s able to drive these scenarios (and more) at scale.
 
Is it possible to handle these without an Inventory Manager (or equivalent)?