© 2022 Black Swan Telecom Journal | • | protecting and growing a robust communications business | • a service of |
Email a colleague |
February 2013
“A frog in a well cannot imagine the ocean.”
Japanese Proverb
For many years now, the business assurance frog has been busy plugging leaks in his back office well. Gosh know he’s been a diligent and tidy fellow. And he’s done a brilliant job at finding money that’s fallen in the cracks between the ordering, provisioning, and billing of telecom services.
But the telecom world does not revolve around “services” alone. In fact, the very language we use to describe telecom operations sometimes confuses our thinking here.
For instance, there are two very distinct kinds of provisioning in a telecom network. “Service provisioning” is the type of provisioning that revenue assurance cares about — the fulfillment of a service to a specific business or consumer.
But there’s also a second kind of provisioning -- “network provisioning” (or network asset management) -- which covers the enormous number of tasks a telecom must do to make fulfilling a service to an individual customer possible. And that includes planning network capacity, ordering network assets, commissioning / decommissioning those assets — even ensuring ample network card spares are available to replace those that fail.
And yet, here’s the funny thing: up to now, all the things that network provisioning does to manage the vast ocean of multi-billion dollar networks are beyond the mission statement of most revenue assurance frogs.
Well, at the recent Telecom Analytics 2013 conference in Atlanta, I ran into John Brooks, Head of Product Management at Subex, who was speaking at the event. And after lunch, John sat down with me and explained his company’s thrust into the network asset management area.
I’d known about Subex’s expertise in network discovery before, but this new approach takes discovery to a higher level of business value than I originally conceived.
Dan Baker: Please set the stage for us, John. What is it about today’s network that makes management of network assets and capital expense more important that it was a couple year ago? |
John Brooks: Tremendous network change from all directions, Dan, is what’s driving the need. In mobile, we’re seeing the GSM and CDMA worlds are migrating into 4G /LTE technologies. In the U.S. market, Sprint, for instance, is now going through the massive process of retiring its IDEN network and ramping up its business to support more 4G.
But change is also occurring in the network backbone too, and here the change is mostly around convergence or consolidation. So, if you look at broadband environments served through, DSL, fiber to the home, copper, HFC/Cable, 3G/4G Mobile, or even a DS3 line serving a business — all are providing the same services through their transport technologies, for instance: IPTV, VoIP, etc. Regardless of your media -- cable, twisted pair copper wire and the mobile edge-- everything is moving to IP transport. And the advantage of this is that if the backbone is transparent, you can connect to any access technology without a problem.
Operators are desperate to know how much more capacity and throughput they can run through pipes and router systems. It’s really a matter of switches and routers becoming bigger, stronger, and faster in the same footprint.
Another big issue is power consumption. Data centers and central offices burn an awful lot of electricity. It costs millions of dollars a month to power up all these systems. So, if you can replace a switch or router with one that carries twice the capacity and uses 50% less power, you can cost justify getting a new switch on power savings alone.
Earlier today at the Telecom Analytics 2013 Conference we heard a talk from experts talking about SDN or Software Defined Networks and they made the point that one of the things that SDN will someday deliver is an ability to power down half your network at night to save power. |
Exactly. And the cloud environment works the same way. By definition, cloud providers (like Amazon, for instance) host services that allow you to throttle up or throttle down resource support, depending on how much demand your users create during a period.
So apply this to a large operator: you’ve got hundreds of devices sitting inside of a big central office and at nighttime, those devices are burning as much power as they were burning during the day but the traffic has dropped off by 87%. What are you doing that for?
Let’s do a deeper dive on network decommissioning. What are some of the complexities around managing device retirement? |
Well, when you think about decommissioning a piece of equipment, there are liable to be thousands of instances of that one model of equipment in the network.
And as you are start retiring assets, often times the equipment will be decommissioned in phases, not all at once. So, there maybe a series of cards inside the switch chassis. And sometimes you’ll decommission the entire chassis; other times you’ll turn down individual cards or components.
And retiring equipment often implies “harvesting” that asset. If you’ve got major markets in New York, Dallas, San Francisco, and Denver, and you’re turning down assets in Philly, you’ve got equipment sparing demands in the other regions, so it makes sense to repopulate the warehouses in those regions with spare equipment.
Likewise, operators need to make some assets available for resale and push them out through the gray market to an operator in South America or maybe a tier 3 operator in the U.S. who needs the equipment being retired.
When you think about the sums of money and investment capital sitting in the telecom network, it kind of staggers the imagination. |
Precisely. And as you’re transitioning equipment in and out of service, time is of the essence because of it high value. You need to carefully “choreograph” your moves to avoid wasteful capital spending and to fully “bleed” your assets of as much value as you possibly can.
Maybe a network box cost an operator $250,000 dollars when purchased and they have 10,000 instances of it across the country: that’s $2.5 billion worth of inventory when they bought it at retail. But two years later the asset may be worth only half the value: $125,000 a piece. So depending on the age of that equipment, the value is going to fall off very, very quickly.
Now if they sit on an idle asset in the network too long, the resale value may go down to $65,000, meaning the 4 months they had that asset sitting idle doing nothing cost a lot of money in lost opportunity. Multiply that across your network and you’re talking tens of millions of dollars.
So you begin the see the value of this equipment choreography. It’s about managing a lot of things: decommissioning, harvesting, or redeploying, bringing assets back into their warehousing, re-inventorying, and gray market selling. And analytics plays a big part in this because that’s what’s going to help predict what the operator should harvest versus what they can go ahead and sell.
So are you are monitoring the spare levels at different locations? |
The actual asset inventories are maintained in existing network inventory systems and warehouse inventory and logistics platforms . Subex works in harmony with those platforms. But our network discovery capability and advanced analytics capabilities allows us to measure the rate of change on the networks. So, we know which parts are moving in and out of the network, where, how fast, and why. These are daily (often hourly) changes to assets that a simple inventory report will not keep pace with.
And one of the keys to our platform’s strength is our off-the-shelf-network adapters. Discovering and understanding network current state at multiple levels is crucial to accurately predicting network needs, which leads directly to capital avoidance. Today we support well over 200 to 300 devices, and the list is being added to constantly. That allows our program to provide real, multi-million dollar capex savings in very short timeframes.
And I understand that your TrueSource network discovery software product is key to enabling Subex’s move into this network provisioning and asset management space. |
The big name for all this is Capacity and Asset Lifecycle Management (CALM) -- basically instrumenting all the steps from planning and budgeting through procurement, receiving, deployment, provisioning, operational use, and retirement.
And yes, the former TrueSource product (now called ROC Data Integrity Management) is the key we have that allows us to discover and gather intelligence on individual network devices that tell us everything from asset and service status to resource consumption for next generation capacity analytics. By using our Capacity Management technology, the customer can glean resource capacity metrics without the need to purchase and maintain a large, expensive network of probes.
Now network capacity planning today is largely being driven by the analysis of traffic. Three years ago, leveraging True Source, Subex teamed up with a few operators to come up with a better way. We don‘t need to look at a single IPDR. Instead we go out and discover network resources and measure the consumption inside of those chasses. We model a variety of resource metrics that tell us about device performance in the face of traffic volumes, to understand how the assets are managing capacity.
If usage data is available, we incorporate that data into device resource modeling for a better view of capacity impacts than usage data alone can‘t offer. For instance, you can pipe through a billion IPDRs and even if the data is metered through due to volume load, it eventually all goes through, with perhaps latency, jitter and even packet loss as a side effect.
Here’s the key: capacity measurement through traffic analytics doesn‘t necessarily tell you how that device has performed. Our method is to say: regardless of how much traffic flowed through, how did that device actually handle the traffic? Frankly speaking, if the resources in that device are overloaded, your quality of service goes down. In short, it really doesn’t matter how much traffic volume was pushed through, it’s how well your network equipment responded to that traffic.
And to do this, we are measuring each device’s vital stats at frequent intervals to see how that node is responding to the demands being put on it.
Wow, John, interrogating the network nodes to deduce capacity data makes a lot of sense. It greatly simplifies the process. Analyzing traffic and gathering network performance data specific to one network device is tricky at best and it certainly is a more complex and expensive method. |
Even still, Dan, though we believe our approach to capacity planning makes the most sense, we can‘t afford to be religious about our own methodology. Every carrier has its own way of measuring capacity via traffic analysis. So using their data is fine because we bring that in all into our platform to ensure we have the broadest range of inputs to analyze. We’ll even incorporate the off-net assets that you need but don’t actually own, if that data is available.
It’s important to note that we are not a reporting mechanism; operators have plenty of those, and if simple reports were effective in lifecycle management, they would be used for that today. The fact is, advanced analytics surround our entire Capacity and Asset Lifecycle Management program.
Being able to produce insights and predictions around key capital avoidance strategies like procurement gating, re-homing , service re-grooming, warehouse sparing levels and policies, service “what-if” analyzers, and asset decommissioning , all from a single end-to-end view across all assets within the lifecycle is crucial for operators to manage and preserve capex. We provide all this actionable intelligence in one holistic ROC (Revenue Operations Center). It is all focused on network capital avoidance.
John, it’s nice to see Subex moving in this direction. I think this sort of work is going to add great value and broaden people’s notions of what business assurance’s mission should be. |
Dan, carriers are in the very early stages of this asset life cycle management area. Most customers we are engaged with today are measuring traffic for network planning purposes.
And what we’re saying to them is “Guys, there’s a whole ecosystem here of which traffic analysis is just the first step.” Of course, the ultimate goal here is capital avoidance and the best use of capital.
Budgets for network capital are constrained, but what if you could make $100 million of network capital feel like a $125 million investment?
So that’s what we’re about: Being able to fully understand your network and how it is evolving and being utilized, applying best in class analytics to drive the most efficient use of existing assets, and ultimately driving significant Capital Avoidance — well into the multi-millions of dollars per year.
Copyright 2013 Black Swan Telecom Journal