Choosing VMware SIOC with Storage Auto Tiering – VCDX Way

Yesterday I wrote a post regarding the IOps, Disk Size, CPU and Memory calculation for the vCOPs Sizing and what would be the impact of it when you submit it for your VCDX defense. Today I will talk about choosing SIOC wisely for your vSphere environment if you have license to use it.

Well, we know many of us working on submitting a design, which has some auto tiering feature in the storage backend. Though it is not mandatory to use any auto tiering technologies and that too it requires special license in the storage subsystem as well, however, when you use it, choose SIOC wisely.

Today I am going to take EMC FAST as an example here. Storage IO Control and EMC FAST VP always complements each other and we should use it where ever possible.

EMC FAST VP manages data placement at the sub-LUN level, thus increasing overall performance. When implemented on a storage system, FAST VP measures, analyzes, and implements a storage-tiering policy much faster and more efficiently than any user could.

Depending on the performance characteristics, FAST VP puts drives of varying performance levels and cost into storage pools. LUNs use the capacity from these pools as needed. FAST VP collects statistics based on I/O activity at 1 GiB slices. These statistics are then used to determine which slices will be promoted to a higher tier of storage. Data that requires a high response time and has a high rate of IOps should be optimized by allowing FAST VP to migrate this data to the SSD drives.

So you can clearly see that it does the balancing act on the LUN level and not on the particular datastore level or VM level and that too it uses a schedule to fire up and takes some time before it can understand the locality reference (hot data). Before that that everything is flat and you may get bombarded with IOps from a particular VM or two.

SIOC is a technology that allows users to configure I/O shares at a datastore level, so that a critical VM receives the I/O throughput and response time when needed, without worrying that another VM on that same datastore won’t hinder the performance of a critical VM. You can from here that SIOC works at the VM level and FAST VP works at the drive level – both complement each other.

So, you may get question from panel why did you choose SIOC or why not. If you choose SIOC with Auto Tiering on your storage, one benefit is immediate effect of storage i/o control from VM disk level, which is really granular. But then the argument is, how operational complex that is? Are the operations guys technical enough to understand the benefit of SIOC? Are they competent enough to argue on a particular VM’s Share and Limit placement? How about the operational effort to manage a properly balanced SIOC environment?

Now, if you don’t choose be prepared to answer to the panelists when they ask you, why did not you choose SIOC? Was that any of the above reason? Was there any cost constraint from the licensing perspective? What would have been different, immediate and long term effect if you would have chosen SIOC precisely in your environment? What would have been effect to your design if you would have chosen only Auto Tiering and not SIOC or SIOC and not any Auto Tiering in the back end storage.

Good luck to all VCDX candidates who are defending in April.

Note: I wrote an extensive article on SIOC and Auto Tiering two years back.


About Prasenjit Sarkar

Prasenjit Sarkar is a Product Manager at Oracle for their Public Cloud with primary focus on Cloud Strategy, Cloud Native Applications and API Platform. His primary focus is driving Oracle’s Cloud Computing business with commercial and public sector customers; helping to shape and deliver on a strategy to build broad use of Oracle’s Infrastructure as a Service (IaaS) offerings such as Compute, Storage, Network & Database as a Service. He is also responsible for developing public/private cloud integration strategies, customer’s Cloud Computing architecture vision, future state architectures, and implementable architecture roadmaps in the context of the public, private, and hybrid cloud computing solutions Oracle can offer.

Leave a Reply