This is the first of a two-part series, you can read part II here.

 

Robin Harris (aka @storagemojo) recently in a blog post asks a question  and thinks solid state devices (SSDs) using SAS or SATA interface in traditional  hard disk drive (HDD) form factors are a bad idea in storage arrays (e.g.  storage systems or appliances). My opinion is that as with many things about storing, processing or moving binary digital data (e.g. 1s and 0s) the  answer is not always clear. That is there may not be a right or wrong answer  instead it depends on the situation, use or perhaps abuse scenario. For some  applications or vendors, adding SSD packaged in HDD form factors to existing  storage systems, arrays and appliances makes perfect sense, likewise for others  it does not, thus it depends (more on that in a bit). While we are talking  about SSD, Ed Haletky (aka @texiwill) recently asked a related question of Fix  the App or Add Hardware, which could easily be morphed into a discussion of Fix  the SSD, or Add Hardware. Hmmm, maybe a future post idea exists there.

 

Lets take a step back for a moment and look at the bigger picture  of what prompts the question of what type of SSD to use where and when along as  well as why various vendors want you to look at things a particular way. There  are many options for using SSD that is packaged in various ways to  meet diverse needs including here and here (see figure 1).

 

Various SSD packaging options
Figure 1: Various packaging and  deployment options for SSD

 

The growing number of startup and established vendors with SSD  enabled storage solutions vying to win your hearts, minds and budget is looking  like the annual NCAA basketball tournament (aka March Madness and march  metrics here and here). Some of vendors have or are adding SSD with SAS or SATA  interfaces that plug into existing enclosures (drive slots). These SSDs have  the same form factor of a 2.5 inch small form factor (SFF) or 3.5 inch HDDs  with a SAS or SATA interface for physical and connectivity interoperability.  Other vendors have added PCIe based SSD cards to their storage systems or  appliances as a cache (read or read and write) or a target device similar to  how these cards are installed in servers.

 

Simply adding SSD either in a drive form factor or as a PCIe card  to a storage system or appliance is only part of a solution. Sure, the hardware  should be faster than a traditional spinning HDD based solution. However, what differentiates  the various approaches and solutions is what is done with the storage systems  or appliances software (aka operating system, storage applications, management,  firmware or micro code).

 

So are SSD based storage systems, arrays and appliances a bad  idea?

 

If you are a startup or established vendor able to start from scratch  with a clean sheet design not having to worry about interoperability and  customer investment protection (technology, people skills, software tools,  etc), then you would want to do something different. For example, leverage off  the shelf components such as a PCIe flash SSD card in an industry standard  server combined with your software for a solution. You could also use extra  DRAM memory in those servers combined with PCIe flash SSD cards perhaps even  with embedded HDDs for a backing or preservation medium.

 

Other approaches might use a mix of DRAM, PCIe flash cards, as  either a cache or target combined with some drive form factor SSDs. In other  words, there is no right or wrong approach; sure, there are different technical  merits that have advantages for various applications or environments. Likewise,  people have preferences particular for technology focused who tend to like one  approach vs. another. Thus, we have many options to leverage, use or abuse.

 

In his post, Robin asks a good question of if nand flash SSD were being put  into a new storage system, why not use the PCIe backplane vs. using nand flash  on DIMM vs. using drive formats, all of which are different packaging options  (Figure 1). Some startups have gone the all backplane approach, some have gone  with the drive form factor, some have gone with a mix and some even using HDDs  in the background. Likewise some traditional storage system and array vendors  who support a mix of SSD and HDD drive form factor devices also leverage PCIe  cards, either as a server-based cache (e.g. EMC VFCahe) or installed as a  performance accelerator module (e.g. NetApp PAM) in their appliances.

 

While most vendors who put SSD drive form factor drives into their  storage systems or appliances (or serves for that matter) use them as data  targets for creating LUNs or file systems, others use them for internal functionality.  By internal functionality I mean instead of the SSD appearing as another drive  or target, they are used exclusively by the storage system or appliance for  caching or similar purposes. On storage systems, this can be to increase the  size of persistent cache such as EMC on the CLARiiON and VNX (e.g. FAST Cache). Another use is on  backup or dedupe target appliances where SSDs are used to store dictionary,  index or meta data repositories as opposed to being a general data pool.

 

Part two of this post looks at the benefits and caveats of SSD in storage arrays.

 

Here are some related links to  learn more about SSD, where and when to use what:
  Why SSD based arrays and storage appliances can be a good idea (Part II)
  IT and storage economics 101,  supply and demand
  Researchers and marketers don't agree on future of nand flash  SSD
  Speaking of speeding up business  with SSD storage
  EMC VFCache respinning SSD and  intelligent caching (Part I)
  EMC VFCache respinning SSD and intelligent caching (Part II)
  SSD options for Virtual (and Physical) Environments: Part I  Spinning up to speed on SSD
  SSD options for Virtual (and Physical) Environments, Part  II: The call to duty, SSD endurance
  SSD options for Virtual (and Physical) Environments Part  III: What type of SSD is best for you?

 

Ok, nuff said for now, check part II.

 

Cheers gs