Earlier this year at TFD 14 in Boston we were given a great run-through of Turbonomic’s new HTML 5 interface. The interface itself breathed new visual life into the product. What I mean by that is while the product was still very functional and extremely useful, it just looked dated. But despite the new interface and features, we had to remember that we were seeing version 5.9. I say this because it felt like a major release.
Today, approximately 5 months later, we are seeing the launch of Turbonomic 6. Previously, the HTML 5 login had to be manually selected. However, going forward, HTML 5 is now the default. The two interfaces are almost completely at feature parity now. The exceptions to this are a handful of one-time creation functions, which still require the traditional UI. The UI however, is but only one of the changes.
THE FUTURE IS CLOUDY
When we saw 5.9, we got a glimpse into how Turbonomic is tackling public cloud. At the time, the big addition was introducing a public cloud cost calculator. With 6.0, the cloud is once again taking another step forward. The initial cost calculations that we saw provided much needed insight into cloud workloads. Now however, the calculations go even deeper.
Let’s start with storage. Whether you are using AWS or Azure, there are plenty of storage options available, all of which vary in price. Most of us would love to jump right into the SSD tier in the hopes of getting the best performance right out of the gate. Some of us might think that going with the cheapest tier is the way to go. Unfortunately most of us have to go by feel when it comes to choosing the right storage tier. With Turbonomic 6.0, performance from storage tiers is now analyzed. This addition will largely take the guess-work out of right-sizing cloud storage.
The other real cool bit about this is that not only will it provide visibility into costs, it will also allow for transitioning between tiers. Say you have an application that gets hammered only part of the time, like a month-end process. Using this you can see when the higher performance is needed and move it into and out of SSD when appropriate.
One other storage-related feature I find noteworthy is the ability to track down orphaned storage instances. This can commonly occur when deleting compute in AWS or Azure. Unless you delete the storage that goes along with those compute resources, you’ll keep paying for them and you may not even realize it. Even today in my on-premises environment I find orphaned VMDKs. I can only imagine how bad this problem can get in larger organizations, especially when talking about cloud.
Along with the storage enhancements, there were improvements made with regards to databases and power management. In terms of databases, real-time scaling is now an option for Azure and AWS. Similar to what we are seeing with moving storage workloads to different tiers, the same philosophy now applies to databases. Do you have an instance that runs some sort of intense roll-up routine once a week? Have it scale up when needed and scale down once it is complete. On the AWS side, there are four metrics used to determine scaling requirements: CPU, memory, IO, and NetThroughput. On the Azure side, the DTU (database transfer unit) is used.
As for power options, there is now the ability to turn off workloads when not in use. Yes, you’ll stay be paying for various resources, such as storage. Where you’ll save money though is in areas such as compute. There are definitely plenty of use cases for this. It could be a matter of scaling up resources for those regularly timed routines. Or maybe if you only run your backup infrastructure at night, you can just power it all down during the day.
There were a slew of other tweaks and features introduced in Turbonomic 6 that I didn’t touch. Rather than just re-hashing the press release or release notes, I wanted to touch on what caught my attention. With cloud technologies being in an ever-growing state, cost and performance management of those resources is only becoming increasingly important. That isn’t to say that private cloud or on-premises datacenters are going to disappear overnight. Turbonomic recognizes that visibility is needed for both of these, whether individually or together. This newest release is a clear indication that they still “get it”. That is to say that they know that folks care about performance and costs, regardless of where the workload actually sits.
You can visit Turbonomic’s website to find out more about 6.0.
Disclaimer: Turbonomic is a sponsor on this site, however I was not requested or paid to write this post, nor was I requested to write this post. Anything written above was on my own accord.