Google touts updates and price cuts to entice users to use cloud for non-bursty stuff
Whether one hears it from Microsoft, Amazon or any other large cloud provider, the message is the same: most people still use public cloud for bursting into occasionally. Google’s latest updates and price cuts seem geared towards incentivising sustained cloud usage, which could bring cloud closer to cost parity with bare-metal. But will enterprises bite?
Google’s senior vice president for technical infrastructure Urs Hölzle announced a slew of updates to the company’s cloud services ranging from storage and compute to the company’s big data-interrogation service, BigQuery.
He said Google would reduce Compute Engine costs by 32 per cent across all instance sizes, regions, and classes, reduce the instance footprint of the company’s PaaS, App Engine, price cloud storage at a consistent ¢2.6 per GB (a 68 per cent drop for most customers), and cut BigQuery on-demand pricing by 85 per cent.
The company also introduced extended OS support in its IaaS platform, and managed VMs for App Engine, which offers more automation when running native code not typically supported within the instance.
“You can start with App Engine and if you ever run into a case where you need more control, or need to use a language or library that App Engine doesn’t support, you can replace part of your application with a VM,” Google wrote on its developer blog.
“Your application may need access to a native resource, such as a file system or network stack; or you might require a library or framework that is only available in C or C++. Current Platform as a service offerings lack this type of support and developers are forced off the cliff into an infrastructure as a service world. With Managed VMs, this is not the case.”
The company also introduced further price cuts for users running VMs for longer periods of time. Users that run VMs in a sustained fashion for a calendar month are entitled to a 25 per cent discount off the total price of each instance.
Above all, the move seems geared towards incentivising more sustained usage of the cloud platform for everything from storage and compute to application development and big data.
But beyond the obvious tactic of staying ahead of companies like Amazon and Microsoft despite being somewhat late to the game, Google’s strategy raises questions about whether enterprises are ready to move steady, predictable workloads over to the cloud.
With some exceptions, public cloud platforms are by most accounts still predominately used for bursting. The reality today is that for most predictable workloads, enterprises can still get a better deal with a bare-metal service: for service providers admin is lower because demands on the hardware and scaling are typically more predictable; there’s little performance tax for the customer because of the elimination of the hypervisor layer; and there’s usually more scope for contract negotiation.
It’s worth pointing out that things are changing on all three of these fronts. But the key question is whether or not they’re changing quickly enough to sway enterprises in the direction public cloud service providers like Google are looking for.
“We’re making it more affordable than it’s ever been before, reintroducing Moore’s Law to the cloud: the cost of virtualized hardware should fall in line with the cost of the underlying real hardware. And you automatically get discounts for sustained use with no long-term contracts, no lock-in, and no upfront costs, so you get the best price and the best performance without needing a PhD in finance,” Hölzle said.