Hi Matt,
TLDR: It's totally happening.
-------
That's a good question. I was the CTO of a leading Fraud and Revenue Assurance company and have witnessed the move with various clients already in a number of situations. Where 10 years the question of cloud-readiness was seldom asked by customers, it has now become a dominant point of discussion when interacting with CSPs. I know of quite a few accounts which have shifted quite a bit of their OSS/BSS to the cloud already. Even legacy applications get outsourced to the cloud in order to get rid of internal infrastructure and resources.
For new initiatives affecting OSS/BSS within the CSP I would say the majority are always considering whether to run this in the cloud and many of them eventually commit to the cloud with these. For existing implementations you see a gradual shift from running systems in-house to the cloud, regardless of whether we are talking shifting legacy applications to the cloud or making the shift as part of a legacy replacement.
I don't know of any meaningful operator today that does not have a cloud strategy in place. They may differ in their approach, but there is no doubt that it is all going in the same direction.
The question here really is what drives this change, as the actual operational cost a lot of the times isn't even that cheap compared to running things in your own data center. It so happens, that the cloud has a LOT to offer (you can finally get rid of Oracle, yeah). I see and have experienced the following main drivers (in no particular order):
- Technically the cloud is now able to keep up with the requirements of the operators. In the early days, that wasn't always so.
- Get rid of infrastructure worries, aging machinery, migration cost.
- Ability to grow with the capabilities of the cloud providers
- Very easy up- and downscaling, especially now that server-less computing is becoming more and more of a thing. Or try AI on-premise with elastic scalability. It would be a huge upfront investment (not to speak of time).
- Ready-made environment support
- Trust in cloud security, which wasn't always like that.
- Ability to take advantage of managed services, so you don't have to do these things yourself.
- The cost can actually be significantly cheaper depending on what exactly you are doing.
- Companies trying to focus on their core business, which is not having to run an IT department like AWS.
- Time-to-market
- More OSS/BSS vendors now offering their solution from within the cloud already (salesforce of course is just an example). But apart from sophisticated OSS/BSS solutions, you can also look at simple tools like JIRA, version control, Microsoft Office packages and so forth. There is really not much point in running, maintaining and administering all of this yourself, if you don't have to. It's a distraction to run this yourself more than anything else.
Lastly, and one of the really important points here: Availability of skilled resources. If you wanted to keep up with a lot of the developments which are incubated in the cloud and/or may only be available in the cloud, you will not be able to hire or train up enough people to replicate this. It's hard enough to keep up with cloud progress already. Not even to speak of the CSPs ability to set up cloud-comparable infrastructure. It's simply not possible, efficient or reasonable. What you need now, is not so much people who know how to run an in-house datacenter, but people who understand how cloud actually works and what it has to offer. Most managers don't fully understand that cloud is a TON more than hardware infrastructure with a little bit of software. I think it's quite the opposite. If you want to be successful utilizing the cloud you should better understand it's capabilities and metrics. Without being too dramatic here you can compare it to playing the piano: If you want it to sound right, you need to know how to play it, not how to build it. So you better have some good cloud-musicians and not someone who knows how to build it (just don't). These musicians unfortunately are a rare species, too, of course.
To your detailed questions:
How fast will the CSPs look to shift from on-premise applications to cloud?Happening already. I think it would be asap. Of course that means something else for each operator.
What teleco's are likely to make the move first? Will they adopt a private or public cloud strategy and why?Tankers take longer to turn. So the big ones will take longer overall. On the other hand they are doing this already and gradually. If you are a small local operator, it may be a lot simpler to go into the cloud with some of your stuff, so it can happen quicker. But it may also not be as rewarding overall.
What pain points are driving this shift (5G, total cost of ownership, new features, other)?See above.
What vendors are best positioned for this switch?Well in my view, the ones which are best positioned are the ones which have an architecture which is natively cloud. You see a lot of legacy vendors which run their stuff in the cloud, but that really just means that they are running old architecture on new cloud architecture, without actually taking advantage of it. It's a little bit like running a 32-bit application on a 64-bit architecture. On Windows this still works, but it will stop working sooner or later.
Customers will also ask questions as to why they still need to pay for DB licenses when it's all running in the cloud etc.
I sometimes wonder why some of the established vendors don't have an initiative going to rebuild their stuff for native cloud. Instead they hold on to their legacy until it's tool late. My prediction is, that many of them will just vanish.
------------------------------
Andrew Tan
layline.io
------------------------------
Original Message:
Sent: May 06, 2019 14:37
From: matt kirk
Subject: Shift to Cloud - Telecom Charging Engine and BSS
Hello,
I was looking to get the group's thoughts on recent commentary around a pending shift to cloud for Telecom charging engines and broader BSS software systems. Google Cloud and AWS have both named telecom IT as a focus point more recently, and there has been some talk of new monetization opportunities from 5G that will require richer applications to harvest.
- How fast will the CSPs look to shift from on-premise applications to cloud?
- What teleco's are likely to make the move first? Will they adopt a private or public cloud strategy and why?
- What pain points are driving this shift (5G, total cost of ownership, new features, other)?
- What vendors are best positioned for this switch?
Would love to hear any / all thoughts on the above and how the industry will or won't make a big push to cloud. Have seen this happen in other vertical software categories, but not yet in any big way in telecom. Wondering if it is inevitable, or telecom is different.
Alternatively, would love to chat in person if any of you will be at the Digital Transformation Conference in Nice later this month.
I have spent a good amount of time on the above and would also be happy to share thoughts / compare notes either in person or on the phone.
Best,
Matt
#DigitalTransformationMaturity
------------------------------
Matt Kirk
Maple Rock Partners
------------------------------