Recently we announced we had joined Google Cloud’s 5G Edge/ISV partner initiative, and that we would be available in Google Cloud with Anthos as a platform for delivering applications and solutions to the network edge.
It’s important to spell out why this is such an important partnership for us and how it will help us help enterprises meet the demands of 5G, IoT, and edge computing, and accelerate their digital transformation.
Why should anyone care?
Anthos is an abstraction layer that allows you to run your containerized applications either on your own hardware, on Google Cloud, on the edge of Google’s cloud, or in AWS, but as part of a single managed deployment. This allows you to avoid becoming overly dependent on a single cloud provider, or being forced to run your own data center because you are afraid of becoming dependent.
Anthos is a non-trivial technical achievement, as cloud providers have long since moved on from providing SSH sessions into boxes and now offer entire ecosystems of hardware, storage, networking, DBaaS, and even lamba services, where you can run code without ever knowing or caring where it’s actually executing.
Only using a single, preferred cloud provider is tremendously seductive to developers but is akin to a giant steel trap with spring-loaded jaws when it comes to the potential for vendor lock in.
VoltDB has always prided itself on being as neutral as it is practical when it comes to public clouds, virtualization, containers, and so on. As a consequence, Anthos’s philosophy is in perfect alignment with our own, which is one of the reasons we are excited about this partnership.
The other main motivation is edge computing. For some people, edge is nothing more than a trendy buzzword. But we operate in a world where latency is critical to everything we do, and being able to locate systems close to our customers is rapidly becoming a key part of our ongoing battle against latency.
The Single Cloud Vendor Conundrum
Very few developers I’ve met are openly enthusiastic about developing and maintaining the skills required to work in on-prem or in multiple public cloud environments. There is a natural tendency to pick one hyperscaler and stick with it.
From the perspective of the developer, this is a brilliant move, as it enables them to reap the synergistic benefits of a single, integrated environment. But it has significant financial implications that can outweigh the short term benefits. It comes down to a tradeoff between time-to-market and long-term costs.
The fundamental problem is that in order to negotiate something you have to have alternative choices available to you.
If you can’t walk away from a negotiation, you aren’t actually a negotiator—you’re a supplicant! If they announce they are doubling prices and it will take you three years to switch, then the actual alternative you face is between paying the new prices forever or the cost of a massive rewrite plus three years of those ludicrous prices.
This is an even bigger problem for publicly quoted companies, as their financials and margins are in the public domain, which means that profitable enterprises may find it very hard indeed to negotiate good terms with their only hyperscaler.
COVID and systemic risk
Another factor to consider is COVID, which has shown society the dangers of over-concentrating key activities in certain geographic or commercial locations.
Here at VoltDB, our specialization in telco and industrial IoT means we’re used to hearing the phrase ‘critical national infrastructure’ in conversations with customers and prospects, and have a practical understanding of what our customers have to do to achieve compliance.
Governments are not in the mood for any more surprises, or, especially, to place blind trust in the reliability claims of people who run our mission-critical energy and emergency systems.
Systemic risk, be it from COVID or ransomware, means that the state will not be taking claims of infallibility from vendors at face value going forward, and within five years we will probably be operating in a notably tougher regulatory regime.
Anthos is not just about economics
While the primary driver behind Anthos is portrayed as operational flexibility, I would argue that the strategic motivation is more economic in nature.
For smaller companies or new startups, Anthos may well be unnecessary, as their goal is to get to market ASAP, which means leveraging a single hyperscaler’s ecosystem as much as they can as part of their journey to a big funding round or an IPO. But in the long term they will need to adopt a multi-cloud strategy to satisfy not just the markets but the regulators. Unless they’ve used something like Anthos to abstract themselves, they will find themselves in the difficult situation of having to port their application to a completely new environment without changing client APIs.
History is full of companies that have tried, and failed, to do this.
What’s next for us and Google?
As our recent product announcement highlighted, VoltDB is clearly focused on solving the increasingly imperative latency challenge presented by 5G and IoT. For tier 1 telcos, addressing this challenge means taking full advantage of cloud native capabilities and edge technology.
VoltDB has been tackling these issues in the telco trenches for years, and the work we’ve done puts us well ahead of the curve. Our partnership with Google is yet another proof point and confirmation that our real-world engagements are helping our customers make the right technology decisions to future-proof their data infrastructures for the new age of telecommunications.
To learn more about VoltDB, start here.