An explanation of IOPS and latency

<I understand this extremely long post is redundant for seasoned storage performance pros – however, these subjects come up so frequently, that I felt compelled to write something. Plus, even the seasoned pros don’t seem to get it sometimes… 🙂 >

IOPS: Possibly the most common measure of storage system performance.

IOPS means Input/Output (operations) Per Second. Seems straightforward. A measure of work vs time (not the same as MB/s, which is actually easier to understand – simply, MegaBytes per Second).

How many of you have seen storage vendors extolling the virtues of their storage by using large IOPS numbers to illustrate a performance advantage?

How many of you decide on storage purchases and base your decisions on those numbers?

However: how many times has a vendor actually specified what they mean when they utter “IOPS”? 🙂

For the impatient, I’ll say this: IOPS numbers by themselves are meaningless and should be treated as such. Without additional metrics such as latency, read vs write % and I/O size (to name a few), an IOPS number is useless.

And now, let’s elaborate… (and, as a refresher regarding the perils of ignoring such things when it comes to sizing, you can always go back here).

 

One hundred billion IOPS…

drevil

I’ve competed with various vendors that promise customers high IOPS numbers. On a small system with under 100 standard 15K RPM spinning disks, a certain three-letter vendor was claiming half a million IOPS. Another, a million. Of course, my customer was impressed, since that was far, far higher than the number I was providing. But what’s reality?

Here, I’ll do one right now: an SSD can do a million IOPS. Maybe even two million.

Go ahead, prove otherwise.

It’s impossible, since there is no standard way to measure IOPS, and the official definition of IOPS (operations per second) does not specify certain extremely important parameters. By doing any sort of I/O test on the box, you are automatically imposing your benchmark’s definition of IOPS for that specific test.

What’s an operation? What kind of operations are there?

It can get complicated.

An I/O operation is simply some kind of work the disk subsystem has to do at the request of a host and/or some internal process. Typically a read or a write, with sub-categories (for instance read, re-read, write, re-write, random, sequential) and a size.

Depending on the operation, its size could range anywhere from bytes to kilobytes to several megabytes.

Now consider the following most assuredly non-comprehensive list of operation types:

  1. A random 4KB read
  2. A random 4KB read followed by more 4KB reads of blocks in logical adjacency to the first
  3. A 512-byte metadata lookup and subsequent update
  4. A 256KB read followed by more 256KB reads of blocks in logical sequence to the first
  5. A 64MB read
  6. A series of random 8KB writes followed by 256KB sequential reads of the same data that was just written
  7. Random 8KB overwrites
  8. Random 32KB reads and writes
  9. Combinations of the above in a single thread
  10. Combinations of the above in multiple threads
…this could go on.

As you can see, there’s a large variety of I/O types, and true multi-host I/O is almost never of a single type. Virtualization further mixes up the I/O patterns, too.

Now here comes the biggest point (if you can remember one thing from this post, this should be it):

No storage system can do the same maximum number of IOPS irrespective of I/O type, latency and size.

Let’s re-iterate:

It is impossible for a storage system to sustain the same peak IOPS number when presented with different I/O types and latency requirements.

Another way to see the limitation…

A gross oversimplification that might help prove the point that the type and size of operation you do matters when it comes to IOPS. Meaning that a system that can do a million 512-byte IOPS can’t necessarily do a million 256K IOPS.

Imagine a bucket, or a shotshell, or whatever container you wish.

Imagine in this container you have either:

  1. A few large balls or…
  2. Many tiny balls
The bucket ultimately contains about the same volume of stuff either way, and it is the major limiting factor. Clearly, you can’t completely fill that same container with the same number of large balls as you can with small balls.
IOPS containers

 

 

 

 

 

 

 

 

 

 

 

 

They kinda look like shotshells, don’t they?

Now imagine the little spheres being forcibly evacuated rapildy out of one end… which takes us to…

Latency matters

So, we’ve established that not all IOPS are the same – but what is of far more significance is latency as it relates to the IOPS.

If you want to read no further – never accept an IOPS number that doesn’t come with latency figures, in addition to the I/O sizes and read/write percentages.

Simply speaking, latency is a measure of how long it takes for a single I/O request to happen from the application’s viewpoint.

In general, when it comes to data storage, high latency is just about the least desirable trait, right up there with poor reliability.

Databases especially are very sensitive with respect to latency – DBs make several kinds of requests that need to be acknowledged quickly (ideally in under 10ms, and writes especially in well under 5ms). In particular, the redo log writes need to be acknowledged almost instantaneously for a heavy-write DB – under 1ms is preferable.

High sustained latency in a mission-critical app can have a nasty compounding effect – if a DB can’t write to its redo log fast enough for a single write, everything stalls until that write can complete, then moves on. However, if it constantly can’t write to its redo log fast enough, the user experience will be unacceptable as requests get piled up – the DB may be a back-end to a very busy web front-end for doing Internet sales, for example. A delay in the DB will make the web front-end also delay, and the company could well lose thousands of customers and millions of dollars while the delay is happening. Some companies could also face penalties if they cannot meet certain SLAs.

On the other hand, applications doing sequential, throughput-driven I/O (like backup or archival) are nowhere near as sensitive to latency (and typically don’t need high IOPS anyway, but rather need high MB/s).

It follows that not all I/O sizes and I/O operations are subject to the same latency requirements.

Here’s an example from an Oracle DB – a system doing about 15,000 IOPS at 25ms latency. Doing more IOPS would be nice but the DB needs the latency to go a lot lower in order to see significantly improved performance – notice the increased IO waits and latency, and that the top event causing the system to wait is I/O:

AWR example Now compare to this system (different format this data but you’ll get the point):

Notice that, in this case, the system is waiting primarily for CPU, not storage.

A significant amount of I/O wait is a good way to determine if storage is an issue (there can be other latencies outside the storage of course – CPU and network are a couple of usual suspects). Even with good latencies, if you see a lot of I/O waits it means that the application would like faster speeds from the storage system.

But this post is not meant to be a DB sizing class. Here’s the important bit that I think is confusing a lot of people and is allowing vendors to get away with unrealistic performance numbers:

It is possible (but not desirable) to have high IOPS and high latency simultaneously.

How? Here’s a, once again, oversimplified example:

Imagine 2 different cars, both with a top speed of 150mph.

  • Car #1 takes 50 seconds to reach 150mph
  • Car #2 takes 200 seconds to reach 150mph

The maximum speed of the two cars is identical.

Does anyone have any doubt as to which car is actually faster? Car #1 indeed feels about 4 times faster than Car #2, even though they both hit the exact same top speed in the end.

Let’s take it an important step further, keeping the car analogy since it’s very relatable to most people (but mostly because I like cars):

  • Car #1 has a maximum speed of 120mph and takes 30 seconds to hit 120mph
  • Car #2 has a maximum speed of 180mph, takes 50 seconds to hit 120mph, and takes 200 seconds to hit 180mph

In this example, Car #2 actually has a much higher top speed than Car #1. Many people, looking at just the top speed, might conclude it’s the faster car.

However, Car #1 reaches its top speed (120mph) far faster than Car # 2 reaches that same top speed of Car #1 (120mph).

Car #2 continues to accelerate (and, eventually, overtakes Car #1), but takes an inordinately long amount of time to hit its top speed of 180mph.

Again – which car do you think would feel faster to its driver?

You know – the feeling of pushing the gas pedal and the car immediately responding with extra speed that can be felt? Without a large delay in that happening?

Which car would get more real-world chances of reaching high speeds in a timely fashion? For instance, overtaking someone quickly and safely?

Which is why car-specific workload benchmarks like the quarter mile were devised: How many seconds does it take to traverse a quarter mile (the workload), and what is the speed once the quarter mile has been reached?

(I fully expect fellow geeks to break out the slide rules and try to prove the numbers wrong, probably factoring in gearing, wind and rolling resistance – it’s just an example to illustrate the difference between throughput and latency, I had no specific cars in mind… really).

And, finally, some more storage-related examples…

Some vendor claims… and the fine print explaining the more plausible scenario beneath each claim:

“Mr. Customer, our box can do a million IOPS!”

512-byte ones, sequentially out of cache.

“Mr. Customer, our box can do a quarter million random 4K IOPS – and not from cache!”

at 50ms latency.

“Mr. Customer, our box can do a quarter million 8K IOPS, not from cache, at 20ms latency!”

but only if you have 1000 threads going in parallel.

“Mr. Customer, our box can do a hundred thousand 4K IOPS, at under 20ms latency!”

but only if you have a single host hitting the storage so the array doesn’t get confused by different I/O from other hosts.

Notice how none of these claims are talking about writes or working set sizes… or the configuration required to support the claim.

What to look for when someone is making a grandiose IOPS claim

Audited validation and a specific workload to be measured against (that includes latency as a metric) both help. I’ll pick on HDS since they habitually show crazy numbers in marketing literature.

For example, from their website:

HDS USP IOPS

 

It’s pretty much the textbook case of unqualified IOPS claims. No information as to the I/O size, reads vs writes, sequential or random, what type of medium the IOPS are coming from, or, of course, the latency…

However, that very same box almost makes 270,000 SPC-1 IOPS with good latency in the audited SPC-1 benchmark:

VSP_SPC1

Last I checked, 270,000 was almost 15 times less than 4,000,000. Don’t get me wrong, 260,000 low-latency IOPS is a great SPC-1 result, but it’s not 4 million SPC-1 IOPS.

Check my previous article on SPC-1 and how to read the results here. And if a vendor is not posting results for a platform – ask why.

Where are the IOPS coming from?

So, when you hear those big numbers, where are they really coming from? Are they just ficticious? Not necessarily. So far, here are just a few of the ways I’ve seen vendors claim IOPS prowess:

  1. What the controller will theoretically do given unlimited back-end resources.
  2. What the controller will do purely from cache.
  3. What a controller that can compress data will do with all zero data.
  4. What the controller will do assuming the data is at the FC port buffers (“huh?” is the right reaction, only one three-letter vendor ever did this so at least it’s not a widespread practice).
  5. What the controller will do given the configuration actually being proposed driving a very specific application workload with a specified latency threshold and real data.
The figures provided by the approaches above are all real, in the context of how the test was done by each vendor and how they define “IOPS”. However, of the (non-exhaustive) options above, which one do you think is the more realistic when it comes to dealing with real application data?

What if someone proves to you a big IOPS number at a PoC or demo?

Proof-of-Concept engagements or demos are great ways to prove performance claims.

But, as with everything, garbage in – garbage out.

If someone shows you IOmeter doing crazy IOPS, use the information in this post to help you at least find out what the exact configuration of the benchmark is. What’s the block size, is it random, sequential, a mix, how many hosts are doing I/O, etc. Is the config being short-stroked? Is it coming all out of cache?

Typically, things like IOmeter can be a good demo but that doesn’t mean the combined I/O of all your applications’ performance follows the same parameters, nor does it mean the few servers hitting the storage at the demo are representative of your server farm with 100x the number of servers. Testing with as close to your application workload as possible is preferred. Don’t assume you can extrapolate – systems don’t always scale linearly.

Factors affecting storage system performance

In real life, you typically won’t have a single host pumping I/O into a storage array. More likely, you will have many hosts doing I/O in parallel. Here are just some of the factors that can affect storage system performance in a major way:

 

  1. Amount of I/O concurrency/threads.
  2. Controller, CPU, memory, interlink counts, speeds and types.
  3. A lot of random writes. This is the big one, since, depending on RAID level, the back-end I/O overhead could be anywhere from 2 I/Os (RAID 10) to 6 I/Os (RAID6) per write, unless some advanced form of write management is employed.
  4. Uniform latency requirements – certain systems will exhibit latency spikes from time to time, even if they’re SSD-based (sometimes especially if they’re SSD-based).
  5. A lot of writes to the same logical disk area. This, even with autotiering systems or giant caches, still results in tremendous load on a rather limited set of disks (whether they be spinning or SSD).
  6. The storage type used and the amount – different types of media have very different performance characteristics, even within the same family (the performance between SSDs can vary wildly, for example).
  7. CDP tools for local protection – sometimes this can result in 3x the I/O to the back-end for the writes.
  8. Copy on First Write snapshot algorithms with heavy write workloads.
  9. Misalignment.
  10. Heavy use of space efficiency techniques such as compression and deduplication.
  11. Heavy reliance on autotiering (resulting in the use of too few disks and/or too many slow disks in an attempt to save costs).
  12. Insufficient cache with respect to the working set coupled with inefficient cache algorithms, too-large cache block size and poor utilization.
  13. Shallow port queue depths.
  14. Inability to properly deal with different kinds of I/O from more than a few hosts.
  15. Inability to recognize per-stream patterns (for example, multiple parallel table scans in a Database).
  16. Inability to intelligently prefetch data.

What you can do to get a solution that will work…

You should work with your storage vendor to figure out, at a minimum, the items in the following list, and, after you’ve done so, go through the sizing with them and see the sizing tools being used in front of you. (You can also refer to this guide).

  1. Applications being used and size of each (and, ideally, performance logs from each app)
  2. Number of servers
  3. Desired backup and replication methods
  4. Random read and write I/O size per app
  5. Sequential read and write I/O size per app
  6. The percentages of read vs write for each app and each I/O type
  7. The working set (amount of data “touched”) per app
  8. Whether features such as thin provisioning, pools, CDP, autotiering, compression, dedupe, snapshots and replication will be utilized, and what overhead they add to the performance
  9. The RAID type (R10 has an impact of 2 I/Os per random write, R5 4 I/Os, R6 6 I/Os – is that being factored?)
  10. The impact of all those things to the overall headroom and performance of the array.

If your vendor is unwilling or unable to do this type of work, or, especially, if they tell you it doesn’t matter and that their box will deliver umpteen billion IOPS – well, at least now you know better 🙂

D

 

 

NetApp vs EMC usability report: malice, stupidity or both?

Most are familiar with Hanlon’s Razor:

Never attribute to malice that which is adequately explained by stupidity.

A variation of that is:

Never attribute to malice that which is adequately explained by stupidity, but don’t rule out malice.

You see, EMC sponsored a study comparing their systems to ones from the company they look up to and try to emulate. The report deals with ease-of-use (and I’ll be the first to admit the current iteration of EMC boxes is far easier to use than in the past and the GUI has some cool stuff in it). I was intrigued, but after reading the official-looking report posted by Chuck Hollis, I wondered who in their right mind will lend it credence, and ignored it since I have a real day job solving actual customer problems and can’t possibly respond to every piece of FUD I see (and I see a lot).

Today I’m sitting in a rather boring meeting so I thought I’d spend a few minutes to show how misguided the document is.

In essence, the document tackles the age-old dilemma of which race car to get by comparing how easy it is to change the oil, and completely ignores the “winning the race with said car” part. My question would be: “which car allows you to win the race more easily and with the least headaches, least cost and least effort?”

And if you think winning a “race” is just about performance, think again.

It is also interesting how the important aspects of efficiency, reliability and performance are not tackled, but I guess this is a “usability” report…

Strange that a company named “Strategic Focus” reduces itself to comparing arrays by measuring the number of mouse clicks. Not sure how this is strategic for customers. They were commissioned by EMC, so maybe EMC considers this strategic.

I’ll show how wrong the document is by pointing at just some of the more glaring issues, but I’ll start by saying a large multinational company has many PB of NetApp boxes around the globe and 3 relaxed guys to manage it all. How’s that for a real example? 🙂

  1. Page 2, section 4, “Methodology”: EMC’s own engineers set up the VNX properly. No mention of who did the NetApp testing, what their qualifications are, and so on. So, first question: “Do these people even know what they’re doing? Have they really used a NetApp system before?”
  2. Page 10, Table A, showing the configurations. A NetApp FAS3070 was used, running the latest code at this moment (8.01). Thanks EMC for the unintended compliment – you see, that system is 2 generations old (the current one is 3270 and the previous one is 3170) yet it can still run the very latest 64-bit ONTAP code just fine. What about the EMC CX3? Can it run FLARE31? Or is that a forklift upgrade? Something to be said for investment protection 🙂
  3. Page 3 table 5-1, #1. Storage pools on all modern arrays would typically be created upfront, so the wording is very misleading. In order to create a new LUN one does NOT NEED to create a pool. Same goes for all vendors.
  4. Same table and section (also mentioned in section 7): Figuring out the space available is as simple as going to the aggregate page, where the space is clearly shown for the aggregates. So, unsure what is meant here.
  5. Regarding LUN creation… Let me ask you a question: After you create a LUN on any array, what do you need to do next? You see, the goal is to attach the LUN to a host, do alignment, partition creation, multipathing and create a filesystem and write stuff to it. You know, use it. NetApp largely automates end-to-end creation of host filesystems and, indeed, does not need an administrator to create a LUN on the array at all. Clearly the person doing the testing is either not aware of this or decided to omit this fact.
  6. Page 4, item 4 (thin provisioning). Asinine statement – plus, any NetApp LUN can be made thick or thin with a single click, whereas a VNX needs to do a migration. Indeed, NetApp does not complicate things whether thin or thick is required, does not differentiate between thin and thick when writing, and therefore does not incur a performance penalty, whereas EMC does (according to EMC documentation).
  7. Page 4, item 5 (Creation of virtual CIFS servers). The Multistore feature is free of charge on all new systems and allows one to create fully segregated, secure multitenancy virtual CIFS, NFS and iSCSI NetApp “partitions” – far beyond the capabilities of EMC. Again, misleading.
  8. Page 4, item 6 (growing storage elements). No measurable difference? Kindly show all the steps to grow a LUN until the new space is visible from the host side. End-to-end is important to real users since they want to use the storage. Or maybe not, for the authors of this document.
  9. Page 5, Item 1. We are really talking here about EMC snapshots? Seriously? Versus NetApp? To earn the right to do so assumes your snapshots are a usable and decent feature and that you can take a good number of them without the box crumbling to pieces. Ask any vendor about a production array with the most snaps and ask to talk to the customer using it. Then compare the number of snaps to a typical NetApp customer’s. Don’t be surprised if one number is a few hundred times less than the other.
  10. Page 5, item 3 (storage tiering): part of a longer conversation but this assumes all arrays need to do tiering. If my solution is optimized to the level that it doesn’t need to do this but yours is not optimized so it needs tiering, why on earth am I being penalized for doing storage more efficiently than you? (AKA the “not invented here” syndrome).
  11. Page 6 item 1 (VMware awareness): NetApp puts all the awareness inside vCenter and, indeed, datastore creation (including volume/LUN/NAS creation and resizing), VM cloning etc. all from within vCenter itself. Ask for a demo and prepare to be amazed.
  12. Page 6, item 2 – (dedupe/compress individual VMs): This one had my blood boiling. You see, EMC cannot even dedupe individual VMs, (impossible, given the fact that current DART code only does dedupe at the file and not block level and no two active VMs will ever be exactly the same), can’t dedupe at all for block storage (maybe in the future but not today), and in general doesn’t recommend compression for VMs! Ask to see the best practices guide that states all this is supported and recommended for active production VMs, and to talk to a customer doing it at scale (not 10 VMs). A feature you can theoretically turn on but that will never work is not quite useful, you see…
  13. Page 8, entire table: Too much to comment on, suffice it to say that NetApp systems come with tools not mentioned in this report that go so far beyond what Unisphere does that it’s not even funny (at no additional cost). Used by customers that have thousands of NetApp systems. That’s how much those tools scale. EMC would need vast portions of the Ionix suite to do anything remotely similar (at $$$). Of course, mentioning that would kinda derail this document… and the piece about support and upgrades is utterly wrong, but I like to keep the surprise for when I do the demos and not share cool IP ideas here 🙂
  14. Page 11, Table B1: In the end, the funniest one of all! If you add up the total number of mouse clicks, NetApp needed 92 vs EMC’s 111. Since the whole point of this usability report is to show overall ease of use by measuring the total number of clicks to do stuff, it’s interesting that they didn’t do a simple total to show who won in the end… 🙂

I could keep going but I need to pay attention to my meeting now since it suddenly became interesting.

Ultimately, when it comes to ease of use, it’s simple to just do a demo and have the customer decide for themselves which approach they like best. Documents such as this one mean less than nothing for actual end users.

I should have another similar list showing clicks and TIME needed to do certain other things. For instance, using RecoverPoint (or any other method), kindly show the number of clicks and time (and disk space) for creating 30 writable clones of a 10TB SQL DB and mounting them on 30 different DB servers simultaneously. Maintaining unique instance names etc. Kinda goes a bit beyond LUN creation, doesn’t it? 🙂

All this BTW doesn’t mean any vendor should rest on their laurels and stop working on improving usability. It’s a never-ending quest. Just stop it with the FUD, please…

Finishing with something funny: Check this video for a good demonstration of something needing few clicks yet not being that easy to do.

Comments welcome.

D

Technorati Tags: , , , ,

EMC conclusively proves that VNX bottlenecks NAS performance

A bit of a controversial title, no?

Allow me to elaborate.

EMC posted a new SPEC SFS result as part of a marketing stunt (which is working, look at what I’m doing – I’m talking about them, if only to clear the air).

In simple terms, EMC got almost 500,000 SPEC SFS NFS IOPS (not to be confused with, say, block-based SPC-1 IOPS) with the following configuration:

  1. Four (4) totally separate VNX arrays, each loaded with SSD storage, utterly unaware of each other (8 total controllers since each box has 2)
  2. Five (5) Celerra VG8 NAS heads/gateways (1 spare), one on top of each VNX box
  3. 2 Control Stations
  4. 8 exported filesystems (2 per VG8 head/VNX system)
  5. Multiple pools of storage (at least 1 per VG8) – not shared among the various boxes, no data mobility between boxes
  6. Only 60TB NAS space with RAID5 (or 15TB per box)

Now, this post is not about whether this configuration is unrealistic and expensive (almost nobody would pay $6m for merely 60TB of NAS, not today). I get it that EMC is trying to publish the best possible number by loading a bunch of separate arrays with SSD. It’s OK as long as everyone understands the details.

My beef has to do with how it’s marketed.

EMC is very vague about the configuration, unless you look at the actual SPEC website. In the marketing materials they just mention VNX, as in “The EMC VNX performed at 497,623 SPECsfs2008_nfs.v3 operations per second”. Kinda like saying it’s OK to take 3 5-year olds and a 6-year old to a bar because their age adds up to 21.

No – the far more accurate statement is “four separate VNXs working independently and utterly unaware of each other did 124,405 SPEC fs2008_nfs.v3 operations per second each“.

All EMC did was add up the result of 4 boxes.

Heck, that’s easy to do!

NetApp already has a result for the 6240 (just 2 controllers doing a respectable 190,675 SPEC NFS ops taking care of NAS and RAID all at once since they’re actually unified, no cornucopia of boxes there) without using Solid State Drives (common SAS drives plus a large cache were used instead – a standard, realistic config we sell every day, and not a “lab queen”).

If all we’re doing is adding up the result of different boxes, simply multiply this by 4 (plus we do have Cluster-Mode for NAS so it would count as a single clustered system with failover etc. among the nodes) and end up with the following result:

  1. 762,700 SPEC SFS NFS operations
  2. 8 exported filesystems
  3. 343TB usable with RAID-DP (thousands of times more resilient than RAID5)

So, which one do you think is the better deal? More speed, 343TB and better protection, or less speed, 60TB and far less protection? 🙂

Customers curious about other systems can do the same multiplication trick for other configs, the sky is the limit!

The other, more serious part, and what prompted me to title the post the way I did, is that EMC’s benchmarking made pretty clear the fact that the VNX is the bottleneck, only able to really support a single VG8 head at top speed, necessitating the need for 4 separate VNX systems to accomplish the final result. So, the fact that a VNX can have up to 8 Celerra heads on top of it means nothing since the back-end is your limiting factor. You might as well stick to a dual-head VG8 config (1 active 1 passive) since that’s all it can comfortably drive (otherwise why benchmark it that way?)

But with only 1 active NAS head you’d be limited to just 256TB max NAS capacity, since that’s how much total space a Celerra head can address as of the time of this writing. Which is probably enough for most people.

I wonder if the NAS heads that can be bought as a package with VNX are slower than VG8 heads, and by how much. You see, most people buying the VNX will be getting the NAS heads that can be packaged with it since it’s cheaper that way. How fast does that go? I’m sure customers would like to know, since that’s what they will typically buy.

I also wonder how fast it would be with RAID6.

Here’s a novel idea: benchmark what customers will actually buy!

So apples-to-apples comparisons can become easier instead of something like this:

Bothapples

For the curious: on the left you see an “Autumn Glory” Malus Floribunda (miniature apple). Photo courtesy of John Fullbright.

D

Technorati Tags: , , , , , , , ,

Questions to ask EMC regarding their new VNX systems…

It’s that time of the year again. The usual websites are busy with news of the upcoming EMC midrange refresh called VNX. And records being broken.

(NEWSFLASH: Watching the webcast now, the record they kept saying they would break ended up being some guy jumping over a bunch of EMC arrays with a motorcycle – and here I was hoping to see some kind of performance record…)

I’m not usually one to rain on anyone’s parade, but I keep seeing the “unified” word a lot, but based on what I’m seeing, it’s all more of the same, albeit with newer CPUs, a different faceplate, and (join the club) SAS. I’m sure the new systems will be faster courtesy of faster CPUs, more RAM and SAS. But are they offering something materially closer to a unified architecture?

Note that I’m not attacking anything in the EMC announcement, merely the continued “unified” claim. I’m sure the new Data Domain, Isilon and Vmax systems are great.

So here are some questions to ask EMC regarding VNX – I’ll keep this as a list instead of a more verbose entry to keep things easy for the ADD-afflicted and allow easier copy-paste into emails 🙂

  1. Let’s say I have a 100TB VNX system. Let’s say I allocate all 100TB to NAS. Then let’s say that all the 100TB is really chewed up in the beginning but after a year my real data requirements are more like 70TB. Can I take that 30TB I’m not using any more and instantly use it for FC? Since it’s “unified” and all? Without breaking best practices for LUN allocation to Celerra? Or is it forever tied to the NAS part and I have to buy all new storage if I don’t want to destroy what’s there and start from scratch?
  2. Is the VNX (or even the NS before it) 3rd-party verified as an over 5-nines system? (I believe the CX is but is the CX/NS combo?)
  3. How is the architecture of these boxes any different than before? It looks like you still have 2 CX SPs, then some NAS gateways. Seems like very much the same overall architecture and there’s (still) nothing unified about it. I call for some truth in advertising! Only the little VNXe seems materially different (not in the software but in the amount of blades it takes to run it all).
  4. Are the new systems licenced by capacity?
  5. Can the new systems use more than the 2TB of FAST Cache?
  6. On the subject of cache, what is the best practice regarding the minimum number of SSDs to use for cache? Is it 8? How many shelves/buses should they be distributed on?
  7. What is the best practice regarding cache oversubscription and how is this sized?
  8. Since the FAST Cache can also cache writes, what are the ramifications if the cache fails? How many customers have had this happen? After all, we are talking about SSDs, and even mirrored SSDs are much less reliable than mirrored RAM.
  9. What’s the granularity for using RecoverPoint to replicate the NAS piece? Seems like it needs to replicate everything NAS as one chunk as a large consistency group, with Celerra Replicator needed for more granular replication.
  10. What’s the granularity for recovering NAS with RecoverPoint? Seems like you can’t do things by file or by volume even. The entire data mover may need to be recovered in one go, regardless of the volumes within.
  11. When using RecoverPoint, does one need to not use storage pools for certain operations? And what does that mean regarding the complexity of implementation?
  12. Speaking of storage pools, when are they recommended, when not, and why? And what does that mean about the complexity of administration?
  13. What functionality does one lose if one does not use pools?
  14. Can one prioritize FAST Cache in pool LUNs or is cache simply on or off for the entire pool?
  15. Can I do a data-in-place upgrade from CX3 or CX4 or is this a forklift upgrade?
  16. Why is FASTv2 not recommended for Exchange 2010 and various other DBs?
  17. If Autotiering is not really applicable to many workloads, what is it really good for?
  18. What is the percentage of flash needed to properly do autotiering on VNX? (it’s only 3% on VMAX since it uses a 7MB page, but VNX uses a 1GB page, which is far more inefficient). Why is FAST still at the grossly inefficient 1GB chunk?
  19. Can FAST on the VNX exclude certain time periods that can confuse the algorithms, like when backups occur?
  20. Is file-level FAST still a separate system?
  21. Why does the low-end VNXe not offer FC?
  22. Can I upgrade from VNXe to VNX?
  23. Does the VNXe offer FAST?
  24. Can a 1GB chunk span RAID groups or is performance limited to 1 RAID group’s worth of drives?
  25. Why are functions like block, NAS and replication still in separate hardware and software?
  26. Why are there still 2 kinds of snapshotting systems?
  27. Are the block snaps finally without a huge write performance impact? How about the NAS snaps?
  28. Are the snaps finally able to be retained for years if needed?
  29. Why are there 4 kinds of replication? (Mirrorview, Celerra Replicator, Recoverpoint, SAN copy)
  30. Why are there still all these OSes to patch? (Win XP in the SPs, Linux on the Control Station and RecoverPoint, DART on the NAS blades, maybe more if they can run Rainfinity and Atmos on the blades as well)
  31. Why still no dedupe for FC and iSCSI?
  32. Why no dedupe for memory and cache?
  33. Why not sub-file dedupe?
  34. Why is Celerra still limited to 256TB per data mover?
  35. Is Celerra still limited to 16TB per volume? Or is yet another, completely separate system (Isilon) needed to do that?
  36. Is Celerra still limited to not being able to share a volume between data movers? Or is, again, Isilon needed to do that?
  37. Can Celerra non-disruptively move CIFS and NFS volumes between data movers?
  38. Why can there not be a single FCoE link to transfer all the protocols if the boxes are “unified”?
  39. Have the thin provisioning performance overheads been fixed?
  40. Have the pool performance bottlenecks been fixed? Or is it still recommended to use normal RAID LUNs for highest performance?
  41. Can one actually stripe/restripe within a FLARE pool now? When adding storage? With thin provisioning?
  42. What is the best practice for expanding, say, a 50 drive pool? How many drives do I have to expand by? Why?
  43. Does one still need to do a migration to use thin provisioning?
  44. Does one need to do yet another migration to “re-thin” a LUN once it gets temporarily chunky?
  45. Have the RAID5 and RAID6 write inefficiencies been fixed? And how?
  46. Will the benchmarks for the new systems use RAID6 or will they, again, show RAID10? After all, most customers don’t deploy RAID10 for everything, and RAID5 is thousands of times less reliable than RAID6. How about some SPC-1 benchmarks?
  47. Why is EMC still not fessing up to using a filesystem for their new pools? Maybe because they keep saying doing so is not a “real” SAN, even in recent communication?
  48. Since EMC is using a filesystem in order to get functionality in the CX SPs like pools, thin provisioning, compression and auto-tiering (and probably dedupe in the future), how are they keeping fragmentation under control? (how the tables have turned!)

What I notice is a lack of thought leadership when it comes to technology innovation – EMC is still playing catch-up with other vendors in many important architectural areas,  and keeps buying companies left and right to plug portfolio holes. All vendors play catch-up to some extent, the trick is finding the one playing catch-up in the fewest areas and leading in the most, with the fewest compromises.

Some areas of NetApp leadership to answer a question in the comments:

  • First Unified architecture (since 2002)
  • First with RAID that has the space efficiency of RAID5, the performance of RAID10 and the reliability of RAID6
  • First with block-level deduplication for all protocols
  • FIrst with zero-impact snapshots
  • First with Megacaches (up to 16TB cache per system possible)
  • First with VMware integration including VM clones
  • First with space- and time-efficient, integrated replication for all protocols
  • First with snapshot-based archive storage (being able to store different versions of your data for years on nearline storage)
  • First with Unified Connect and FCoE – single cable capability for all protocols (FC, iSCSI, NFS, CIFS)

However, EMC is strong when it comes to marketing, messaging and – wait for it – the management part. Since it’s amazingly difficult to integrate all the technologies EMC has acquired over the years (heck, it’s taking NetApp forever to properly integrate Spinnaker and that’s just one other architecture), EMC is focusing instead on the management of the various bits (the current approach being Unisphere, tying together a subset of EMC’s acquisitions).

So, Unified Storage in EMC-speak really means unified management. Which would be fine if they were upfront about it. Somehow, “our new arrays with unified management but not unified architecture” doesn’t quite roll off the tongue as easily as “unified storage”.

Mike Riley eloquently explains whether it’s easier to fix an architecture or fix management here. Ultimately, unified management can’t tackle all the underlying problems and limitations, but it does allow for some very nice demos.

A cool GUI with frankenstorage behind it is like putting lipstick on a pig, or putting a nice shell on top of a car cobbled together from disparate bits. The underlying build is masked superficially, until it’s not… usually, at the worst possible time.

Sure, ultimately, management is what the end user interfaces with. Many people won’t really care about what goes on inside, nor have the time or inclination to learn. I merely invite them to start thinking more about the inner bits, because when things get tricky is also when something like a portal GUI meshing 4-5 different products together also stops working as expected, and that’s also when you start bouncing between 3-4 completely different support teams all trying to figure out which of the underlying products is causing the problem.

Always think in terms of what happens if something goes wrong with a certain subsystem and always assume things will break – only then can you have proper procedures and be prepared for the worst.

 

 

 

 

 

 

 

 

And always remember that the more complex a machine, the more difficult it can be to troubleshoot and fix when it does break (and it will break – everything does). There’s no substitute for clean and simple engineering.

Of course, Rube Goldberg-esque machines can be entertaining… if entertainment is what you’re after 🙂

D

 

Technorati Tags: , , , , , , , , , , ,

 

FUD tales from the blogosphere: when vendors attack (and a wee bit on expanding and balancing RAID groups)

Haven’t blogged in a while, way too busy. Against my better judgment, I thought I’d respond to some comments I’ve seen on the blogosphere, adding one of my trademark extremely long titles. Part response, part tutorial. People with no time to read it all: Skip to the end and see if you know the answer to the question or if you have ideas on how to do such a thing.

It’s funny how some vendors won’t hesitate to wholeheartedly agree when some “independent” blogger criticizes their competition (before I get flamed, independent in quotes since, as I discussed before, there ain’t no such thing whether said blogger realizes it or not – being biased is a basic human condition).

The equivalent of someone posting in an Audi forum about excessive brake dust, and having guys from Mercedes and BMW chime in and claim how they “tested” Audis and indeed they had issues (but of course!) and how their cars are better now and indeed maybe Audi doesn’t have as much of a lead any more (if, indeed, they ever did). I think the term for that is “shill” but I can understand taking every opportunity to harm an opponent.

So the “Storage Architect” posted entries asking about certain features to be implemented on NetApp storage, one of them being able to reduce the size of an aggregate. Then everyone and their mum jumped on and complained how on earth such an important feature isn’t there 🙂 BTW I’m not saying such a thing wouldn’t be useful to have from time to time. I’ll just try to explain why it’s tricky to implement and maybe ways to avoid problems.

For the uninitiated, a NetApp aggregate is a collection of RAID-DP RAID groups, that are pooled, striped and I/O then hits all the drives from all RAID groups equally for performance. You then carve out volumes out of that aggregate (containers for NFS, CIFS, iSCSI, FC).

A pretty simple structure, really, but effective. Similar constructs are used by many other storage vendors that allow pooling.

So, the question was, why not be able to make an aggregate smaller? (you can already make it bigger on-the-fly, as well as grow or shrink the existing volumes within).

An HP guy them proceeded to complain about how he put too few drives in an aggregate and ended up with an imbalanced configuration while trying to test a NetApp box.

So, some basics:  the following picture shows a well-balanced pool – notice the equal number of drives per RAID group:

The idea being that everything is load-balanced:

Makes sense, right?

You then end up with pieces of data across all disks, which is the intent. Growing it is easy – which is, after all, what 99.99% of customers ever want to do.

However, the HP dude didn’t have enough disks to create a balanced config with the default-sized RAID group (16). So he ended up with something like this, not performance-optimal:

So what the HP dude wanted to do, was to reduce the size of the RAID group and remove drives, even though he expanded the aggregate (and by extension the RAID group) originally.

Normally, before one starts creating pools of storage (with any storage system), one also knows (or should) what one has to play with in order to get the best overall config. It’s like “I want to build a 12-cylinder car engine, but I only have 9 cylinders”. Well – either buy more cylinders, or build an 8-cylinder engine! Don’t start building the 12-cylinder engine and go “oops” 🙂 This is just Storage 101. Mistakes can and do happen, of course.

So, with the current state of tech, if I only had 20 drives to play with (and no option to get more), assuming no spares, I’d rather do one of the following:

  1. Aggregate with 10 + 10 RAID groups inside or
  2. Use all 20 drives in a single RAID group for max space
  3. Ask someone that knows the system better than I do for some advice

This is common sense and both doable and trivial with a NetApp system. The idea is you set the desired RAID group size for that aggregate BEFORE you put in disks. Not really difficult and pretty logical.

For instance, aggr options HPdudeAggr raidsize 10 before adding the drives would have achieved #1 above. Graphically, the Web GUI has that option in there as well, when you modify an aggregate. The option exists and it’s well-known and documented. Not knowing about it is a basic education issue. Arguing that no education should be needed to use a storage device (with an extreme number of features) properly even for deeply involved, low-level operations, is a romantic notion at best. Maybe some day. We are all working hard to make it a reality. Indeed, a lot of things that would take a really long time in the past (or still, with other boxes) have become trivialized – look at SnapDrive and the SnapManager products, for instance.

Back to our example: if, in the future, 10 more disks were purchased, and approach #1 above was taken, one would simply add the ten disks to the aggregate with aggr add HPdudeAggr 10. Resulting in a 10+10+10 config.

But what if I had done #2 above (make a 20-drive RAID group the default for that aggregate)?

Then, simply, you’d end up imbalanced again, with a 20+10. Some thought is needed before embarking on such journeys.

Maybe a better approach would be to add, say, a more reasonable number of drives to achieve good balance? Adding 12 more drives, for example, would allow for an aggregate with 16+16 drives. So, one could simply change the raidsize using aggr options HPdudeAggr raidsize 16, then, add the 12 disks to the aggregate with aggr add HPdudeAggr -g all 12.

This would expand both RAID groups contained within the aggregate dynamically to 16 drives per, resulting in a 16+16 configuration. Which, BTW, is not something you can easily do with most other storage systems!

Having said all that, I think that for people that are not storage savvy (or for the storage savvy that are suffering from temporary brain fog), a good enhancement would be for the interfaces to warn you about imbalanced final configs and show you what will be created in a nice graphical fashion, asking you if you agree (and possibly providing hints on how it could be done better).

I’m not aware of any other storage system that does that degree of handholding but hey, I don’t know everything.

Indeed, maybe the nature of the other posts was being bait so I’ll obligingly take the bait and ask the question so you can advertise your wares here: 🙂

Is anyone aware of a well-featured storage system from an established, viable vendor that currently (Aug 7, 2010, not roadmap or “Real Soon Now”) allows the creation of a wide-striped pool of drives with some RAID structures underneath; then allows one to evacuate and then destroy some of those underlying RAID groups selectively, non-disruptively, without losing data, even though they already contain parts of the stripes; then change the RAID layout to something else using those same existing drives and restripe without requiring some sort of data migration to another pool and without needing to buy more drives? Again, NOT for expansion, but for the shrinking of the pool?

To clarify even further: What the HP guy did was exactly this: He had 20 drives to play with, he created by mistake a pool with 2 RAID groups, 14+2 and a 2+2, how would your solution take those 2 RAID groups, with data, and change the config to something like 10 + 10 without needing more drives or the destruction of anything?

Can you dynamically reduce a RAID group? (NetApp can dynamically expand, but not reduce a RAID group).

I’m not implying such a thing doesn’t exist, I’m merely curious. I could see ways to make this work by virtualizing RAID further. Still, it’s just one (small) part of the storage puzzle.

The one without sin may cast the first stone! 🙂

D

Technorati Tags: ,,