Can RAID extend the life of nand flash SSD?

Storage I/O trends

Can RAID extend nand flash SSD life?

Imho, the short answer is YES, under some circumstances.

There is a myth and some FUD that RAID (Redundant Array of Independent Disks) can shorten the life durability of nand flash SSD (Solid State Device) vs. HDD (Hard Disk Drives) due to extra IOP’s. The reality is that depending on how configured, RAID level, implementation and other factors, nand flash SSD can be extended as I discuss in this here video.

Video

Nand flash SSD cells and wear

First, there is a myth that nand flash SSD does not have moving parts like hard disk drives (HDD’s) thus do not wear out or break. That is just a myth in that nand flash by its nature wears out with write usage. This is due to how they store data in cells that have a rated number of program erase (P/E) cycles that vary by type of medium. For example, Single Level Cell (SLC) has a longer P/E life duration vs. Multi-Level Cells (MLC) and eMLC that stack multiple cells together.

There are a number of factors that contribute to nand flash wear, also known as duty cycle or durability tied to P/E. For example, some storage systems or controllers do a better job both at the lower level flash translation layer (FTL) in addition to controllers, firmware, caching using DRAM and IO optimization such as write ordering or grouping.

Now what about this RAID and SSD thing?

Ok first as a recap keep in mind that there are many RAID levels along with variations, enhancements and where, or how implemented ranging from software to hardware, adapters to controllers to storage systems.

In the case of RAID 1 or mirroring, just like replication or other one to one or one too many copy operation a write to one device is echoed to another. In the case of RAID 5, data is spread across drives and parity; however, the parity is rotated across all drives in an equal manner.

Some FUD or myths or misunderstandings come into play is that not all RAID 5 implementations as an example are not the same. Some do a better job of buffering or caching data in battery protected mirrored DRAM memory until a full stripe write can occur, or if needed, a partial write.

Another attribute is the chunk or shard size (how much data is sent to each drive member) along with the stripe width (how many drives). Some systems have narrow stripes of say 3+1 or 4+1 or 5+1 while others can be 14+1 or 15+1 or wider. Thus, data can be written across a wider number of drives reducing the P/E consumption or use of a single drive depending on implementation.

How about RAID 6 (dual parity)?

Same thing, it is a matter of how well the implementation is, how the write gathering is done and so forth.

What about RAID wearing out nand flash SSD?

While it is possible that it has or can occur depending on type of RAID implementation, lack of caching or optimization, configuration, type of SSD, RAID level and other things, in general I will say myth busted.

Want some proof?

I could go through a long technical proof point and citing lots of facts, figures, experts and so forth leaving you all silenced and dazed similar to the students listening to Ben Stein in Ferris Buelers day off (Click here to see what I mean) asking “anybody anybody Buleler?

Ben Stein via https://nostagjicmoviesandthings.blogspot.com
Image via nostagjicmoviesandthings.blogspot.com

How about some simple SSD and storage math?

On a very conservative basis, my estimate is that around 250PB of nand flash SSD drives are shipped and installed on a revenue basis attached to or in storage systems and appliances. Combine what Dell + DotHill + EMC + Fujitsu + HDS + HP + IBM (including TMS) + NEC + NetApp + NEC + Oracle among other legacy along with new all flash as well as hybrid vendors (e.g. Cloudbyte, FusionIO (Via their Nexgen acquisition), Kaminario, Greenbytes, Nutanix or Nimble, Purestorage, Starboard or Solidfire, Tegile or Tintri, Violin or Whiptail among others).

It is also a safe assumption based on how customers configure and use those and other storage systems is with some form of RAID. Thus if things were as bad as some researchers were, vendors and their pundits have made them out to be, wouldn’t’t we be hearing of those issues?

Is it just a RAID 5 problem and that RAID 6 magically corrects the problem?

Well, that depends on apples to apples vs. apples to oranges comparisons.

For example if you are using a 14+2 (16 drive) RAID 6 to compare to say a 3+1 (4 drive) RAID 5 that is not a fair comparison. Granted, it is a handy one if you are a vendor that supports wider RAID groups, stripes and ranks vs. those who do not. However also keep in mind that some legacy vendors actually also support wide stripes and RAID groups.

So in some cases the magic is not in the RAID level, rather the implementation or how configured or lack thereof.

Video

Watch this TechTarget produced video recorded live while I was at EMCworld 2013 to learn more.

Otherwise, ok, nuff said (for now).

Cheers
Gs

Greg Schulz – Author Cloud and Virtual Data Storage Networking (CRC Press), The Green and Virtual Data Center (CRC Press) and Resilient Storage Networks (Elsevier)
twitter @storageio

All Comments, (C) and (TM) belong to their owners/posters, Other content (C) Copyright 2006-2024 Server StorageIO and UnlimitedIO LLC All Rights Reserved

Have SSDs been unsuccessful with storage arrays (with poll)?

Storage I/O Industry Trends and Perspectives

I hear people talking about how Solid State Devices (SSDs) have not been successful with or for vendors of storage arrays, particular legacy storage systems. Some people have also asserted that large storage arrays are dead at the hands of new purpose-built SSD appliances or storage systems (read more here).

As a reference, legacy storage systems include those from EMC (VMAX and VNX), IBM (DS8000, DCS3700, XIV, and V7000), and NetApp FAS along with those from Dell, Fujitsu, HDS, HP, NEC and Oracle among others.

Granted EMC have launched new SSD based solutions in addition to buying startup eXtremeIO (aka Project X), and IBM bought SSD industry veteran TMS. IMHO, neither of those actions by either vendor signals an early retirement for their legacy storage solutions, instead opening up new markets giving customers more options for addressing data center and IO performance challenges. Keep in mind that the best IO is the one that you do not have to do with the second best being the least impact to applications in a cost-effective way.

SSD, IO, memory and storage hirearchy

Sometimes I even hear people citing or using some other person or source to attribute or make their assertions sound authoritative. You know the game, according to XYZ or, ABC said blah blah blah blah. Of course if you say or repeat something often enough, or hear it again and again, it can become self-convincing (e.g. industry adoption vs. customer deployments). Likewise depending on how many degrees of separation exists between you and the information you get, the more that it can change from what it originally was.

So what about it, has SSD not been successful for legacy storage system vendors and is the only place that SSD has had success is with startups or non-array based solutions?

While there have been some storage systems (arrays and appliances) that may not perform up to their claimed capabilities due to various internal architecture or implementation bottlenecks. For the most part the large vendors including EMC, HP, HDS, IBM, NetApp and Oracle have done very well shipping SSD drives in their solutions. Likewise some of the clean sheet new design based startup systems, as well as some of the startups with hybrid solutions combing HDDs  and SSDs have done well while others are still emerging.

Where SSD can be used and options

This could also be an example where myth becomes reality based on industry adoption vs. customer deployment. What this means is that the myth is that it is the startups that are having success vs. the legacy vendors from an industry adoption conversation standpoint and thus believed by some.

On the other hand, the myth is that vendors such as EMC or NetApp have not had success with their arrays and SSD yet their customer deployments prove otherwise. There is also a myth that only PCIe based SSD can be of value and that drive based SSDs are not worth using which I have a good idea where that myth comes from.

IMHO it is a depends, however safe to say from what I have seen directly that there are some vendors of storage arrays, including so-called legacy systems that have had very good success with SSD. Likewise have seen where some startups have done ok with their new clean sheet designs, including EMC (Project X). Oh, at least for now I am not a believer that with the all SSD based project “X” over at EMC that the venerable VMAX  formerly known as DMX and its predecessors Symmetric have finally hit the end of the line. Rather they will be positioned and play to different markets for some time yet.

Over at IBM I don’t think the DS8000 or XIV or V7000 and SVC folks are winding things down now that they bought SSD vendor TMS who has SSD appliances and PCIe cards. Rest assured there have been success by PCIe flash card vendors both as targets (FusionIO) and cache or hybrid cache and target systems such as those from Intel, LSI, Micron, and TMS (now IBM) among others. Oh, and if you have not noticed, check out what Qlogic, Emulex and some of the other traditional HBA vendors have done with and around SSD caching.

So where does the FUD that storage systems have not had success with SSD come from?

I suspect from those who would rather not see or hear about those who have had success taking away attention from them or their markets. In other words, using Fear, Uncertainty and Doubt (FUD) or some community peer pressure, there is a belief by some that if you hear enough times that something is dead or not of a benefit; you will look at the alternatives.

Care to guess what the preferred alternative is for some? If you guessed a PCIe card or SSD based appliance from your favorite startup that would be a fair assumption.

On the other hand, my educated guess (ok, its much more informed than a guess ;) ) is that if you ask a vendor such as EMC or NetApp they would disagree, while at the same time articulate benefits of different approaches and tools. Likewise, my educated guess is that if you ask some others, they will say mixed things and of course if you talk with the pure plays, take a wild yet educated guess what they will say.

Here is my point.

SSD, DRAM, PCM and storage adoption timeline

The SSD market, including DRAM, nand flash (SLC or MLC or any other xLC), emerging PCM or future mram among other technologies and packaging options is still in its relative infancy. Yes, I know there have been significant industry adoption and many early customer deployments, however talking with IT organizations of all size as well as with vendors and vars, customer deployment of SSD is far from reaching its full potential meaning a bright future.

Simply putting an SSD, card or drive into a solution does not guarantee results.

Likewise having a new architecture does not guarantee things will be faster.

Fast storage systems need fast devices (HDD, HHDD and SSDs) along with fast interfaces to connect with fast servers. Put a fast HDD, HHDD or SSD into a storage system that has bottlenecks (hardware, software, architectural design) and you may not see the full potential of the technology. Likewise put fast ports or interfaces on a storage system that has fast devices however also a bottleneck in its controller has or system architecture and you will not realize the full potential of that solution.

This is not unique to legacy or traditional storage systems, arrays or appliances as it is also the case with new clean sheet designs.

There are many new solutions that are or should be as fast as their touted marketing stories present, however just because something looks impressive in a YouTube video or slide deck or WebEx does not mean it will be fast in your environment. Some of these new design SSD based solutions will displace some legacy storage systems or arrays while many others will find new opportunities. Similar to how previous generation SSD storage appliances found roles complementing traditional storage systems, so to will many of these new generation of products.

What this all means is to navigate your way through the various marketing and architecture debates, benchmarks battles, claims and counter claims to understand what fits your needs and requires.

StorageIO industry trends cloud, virtualization and big data

What say you?

Ok, nuff said

Cheers
Gs

Greg Schulz – Author Cloud and Virtual Data Storage Networking (CRC Press, 2011), The Green and Virtual Data Center (CRC Press, 2009), and Resilient Storage Networks (Elsevier, 2004)

twitter @storageio

All Comments, (C) and (TM) belong to their owners/posters, Other content (C) Copyright 2006-2012 StorageIO and UnlimitedIO All Rights Reserved

SAS Disk Drives Appearing in Larger Mid-Range Arrays

Storage I/O trends

2.5″ and 3.5″ 10,000RPM (10K) and 15,000RPM (15K) SAS disk drives have been available for entry level and SMB solutions from the likes of Dell, EMC, HP, IBM, Infotrend, LSI, NetApp, Promise, Sun and Xyratex among others for some time now. HDS recently introduced a new mid-range solution the AMS 2000 that supports up to 480 3G SAS drives in place of traditional 2G or 4G Fibre Channel disk drives.

The benefit of supporting SAS disk drives moving forward is that as volume production and adoption increases, price will decline making the drives more affordable not to mention that the controller and interface chip sets and internal adapters support both SAS and SATA disk drives removing additional cost and complexity from storage systems. Similar to how Fibre Channel disk drives replaced parallel SCSI, SSA or other propriety disk drives, and how SATA replaced parallel ATA (PATA) drives, dual ported SAS disk drives continue the cycle of replacing Fibre Channel disk drives at the entry level and mid-range and eventually finding their way into high-end and ultra-high-end storage arrays over the next couple of years as 6G SAS drives begin to appear.

Watch for a SAS drive to appear in a storage system or server near you soon.

Ok, nuff said.

Cheers gs

Greg Schulz – Author Cloud and Virtual Data Storage Networking (CRC Press), The Green and Virtual Data Center (CRC Press) and Resilient Storage Networks (Elsevier)
twitter @storageio

All Comments, (C) and (TM) belong to their owners/posters, Other content (C) Copyright 2006-2024 Server StorageIO and UnlimitedIO LLC All Rights Reserved