Is your ISCSI fast enough?

iSCSI is a great technology. It gives you the ability to create SANs very cheaply and easily, without having to become a guru in fibre channel or put yourself into deep debt buying all the fibre equipment. By using easily available networking equipment you can add storage to existing boxes, even if you want to go crazy and do shared access, clustering or other high-end features.

A lot of vendors provide basically free iSCSI targets (there’s even one in Windows Server these days), and almost every OS has a free software initiator to connect to those targets. Yes, we can bang on about whether software or hardware initiators are better, but software are free and work so well that most iSCSI hardware initiator vendors have stopped bothering.

For the uninitiated, a target is the machine with the storage, and an initiator is the machine connecting across the network to that storage – simple (if you say it quickly but in reality it’s not that hard these days).

So now that I’ve oversimplified iSCSI, let’s look at it’s performance. It’s pretty good for most things, but in my experience suffers in the random read/write area – small read and writes carry quite a bit of network overhead etc.

So what can be done about that?

Hmmm … spend millions with a big-name vendor on a super-duper iSCSI target that you don’t need, or … cache the random reads and writes at the initiator end, so that those small, frequently used blocks don’t actually travel back and forth across the iSCSI network.

That sounds very attractive, and in fact it is very easy to achieve. Adaptec has a product called “maxCache Plus” – which is caching and tiering software, and is built into all 8 Series “Q” controllers (eg 8885Q). Now prior to this version of the software that works with these controllers, caching only worked on devices attached to the controller, which an iSCSI target definitely isn’t. However now with “maxCache Plus”, you can cache or tier any storage in your server, including iSCSI targets that are in fact disk drives sitting somewhere out on the network, but just appear to be drives inside the server.

So …

Plug in an 8Q controller, connect 1 or more SSDs (you don’t even need to make them RAID – you can use them as single raw disks), then add caching to an exiting logical device (disk) – using what we call “CachedLD” (Cached Logical Device).

That logical device can be an iSCSI target, an on-board RAID array off the motherboard, or even a RAID array from a competing RAID card vendor … and did I mention that you don’t need to reconfigure the existing data on the server, or make any changes to the server configuration? That sounds too good to be true, but in fact it works seamlessly.

So you could in fact make your own iSCSI target using cheap hardware and pretty much free software to create a very cheap network storage solution – then accelerate the random reads and writes to that locally in your server, rather than spending a fortune in the iSCSI arena. Hopefully you’ll use an Adaptec RAID card in that iSCSI box, but more than likely you’ll use free Linux software RAID – what the heck, just do it :-)

By the way … what happens if the SSD fails (in the case of using a single SSD) – absolutely nothing to your data – you just go back to original performance before installing caching – CachedLD cannot hurt your data.

Now that’s something to think about.

Ciao
Neil

 

Posted in Advisor - Neil, Application Environments, General, Platforms, Storage Applications, Storage Interconnects & RAID, Storage Management | Leave a comment

Playing games beats working … any day …

Was recently asked a question about SCSI-Express by a customer … eg when are we going to have controller out to handle such devices etc. So I flicked back a question … what devices?

He quick threw back some links of Asus and Gigabyte motherboards that have SCSI-Express connectors on them – and yes of course they are gaming boards (http://www.gigabyte.co.nz/products/product-page.aspx?pid=4959#ov).

I dug up an old link from the STA (http://www.scsita.org/library/documents/STA_TechShowcase_DataSheet_AdvanPCIe-print.pdf) and all you’ll find all over the wording is “enterprise”. If it had been written today, it would have the word “datacenter” stuffed in there as well because that’s the new marketing buzzword (especially “Hyperscale Datacenters”), but I guess that document must be more than 10 minutes old which makes it old hat in this world’s marketing analysis.

However, it made me think … who is going to use SCSI-Express and/or NVMe devices? Are they going to be so much cheaper than the current crop of straight PCIe NVRAM cards (eg LSI Nitro or FusionIO) that everyone will run out and buy one? When was the last time that people put FusionIO cards in gaming machines?

Maybe I’m missing the point. Maybe the gamers of the world do indeed spend that much money on storage. Somehow I thought they spent all the money on processors, video cards, memory and fish-tank cooling systems … not PCIe SSD.

Now I’m not saying what we are doing in this department because heaven forbid, that would get me in hot water, so you can just wait for the marketing team to talk about “real world” stuff, but seriously, am I getting too old?

I’m confused … is there a “gaming enterprise” segment of this market that I hadn’t heard about.

Ciao
Neil

Posted in Advisor - Neil, Application Environments, General, Platforms, Storage Applications, Storage Interconnects & RAID, Storage Management | Leave a comment

Content delivery …

We have a constant battle in the halls of PMC … what exactly should the blog look like? While some of that discussion is based around the look and feel, most of it is based around the content.

Should it be technical? … a real propeller-head’s delight?
Should it be generic in nature (in my opinion boring as …)
Should it be perceptive and propose the future (I can hear legal fainting from here)
OR
Should it be a combination of all these things?

I think it should have lots more pictures of Ducati motorbikes (my pride and joy), but the marketing team don’t really agree on this. So here’s your chance … tell me what it is you want (that’s dangerous I know) … and if it’s physically possible I’ll give it a go.

Meanwhile I’m going to sit back and look at the bike because it’s way too cold to ride the thing at the moment :-)

Ciao
Neil

Posted in Advisor - Neil, Application Environments, General, Platforms, Storage Applications, Storage Interconnects & RAID, Storage Management | Leave a comment

So who is using whitebox servers?

We have this discussion all the time at Adaptec … who is using whitebox servers?

Focus number 1 these days is datacenters – and yes a large portion of them use whitebox. Once they get to the numbers that those guys use it makes sense to have them built for you by an ODM in Taiwan etc – the brand name players get a bit expensive in those numbers.

But at the other end of the spectrum, I recently came across a group of organisations who also promote and use whitebox servers and I must say I was a little surprised.

Managed Service providers. A lot of these guys provide total solutions to small to medium business (and some pretty large business as well) … they do the telecoms, the networking, the internet and these days are pushing to provide the hardware for servers and workstations either installed in the customer site with remote management, or servers sitting in their own small/medium-sized datacenters.

So why do they want whitebox? Mainly because they don’t want to have to call the brand name hardware vendor every time they think they have a hardware problem, and they don’t want that brand name player coming to the customer and promoting their own services (lots of competition in this market).

So it’s better to use whitebox (generally cheaper) – they have control over the build of the box to suit the customer, and they have control over all components from software to hardware, and can provide the service they want to a customer without bringing in a third-party brand-name vendor to replace a hard drive etc.

Makes sense to me.

So who else is using whitebox out there and what are you using them for?

Ciao
Neil

Posted in Advisor - Neil, Application Environments, General, Platforms, Storage Applications, Storage Interconnects & RAID, Storage Management | Leave a comment

Thinking outside the square … or “how to build a small server”

Everyone today likes SSDs, but no-one wants to buy more than they need and more than they can afford. The following article shows you some options in how to build a server, small or large, using the benefits of SSD without spending a fortune. Note that the Hybrid RAID option talked about here is available on all of Adaptec’s RAID controllers from the 6405E upwards.

The underlying basis for this article is Hybrid RAID. This is unique to Adaptec controllers and simply means then when the controller finds a mirror or RAID 10 made of both SSD and HDD, it will write data to both drives (as in the traditional mirror method), but will direct all reads to the SSD, ignoring the HDD for reads. Since the one major benefit that SSD brings to the table is read speed, it makes sense to read from the fast devices, rather than trying to balance read between the devices.

This is automatic – the controller will just do this by itself when it finds the configs below, so the user does not even have to think about anything unusual in the configuration apart from the physical devices used.

So why this article? I found this doc that I wrote ages ago, and only just worked out how to put images in this blog :-)

Option 1 – Large Capacity Servers

How to make maximum use of space in a large capacity server, while still gaining a fast-booting server (to minimize downtime).
Scenario: 8-bay server. Customer wants to gain as much space as possible while still retaining fast boot time. Customer is conservative and wishes to keep OS away from data (OS in mirror and data in RAID 5 for maximum capacity) – or more likely they are stuck with the limitations of an OS that can’t handle a boot drive of more than 2TB :-)

Traditional method

The problem with this method is that is wastes two bays in the server to the OS boot drives. Therefore capacity is limited to 6 drives in R5 (N-1 means 5 drives data capacity). The server does not boot particularly quickly as it is booting from standard HDD in mirror (not a very fast form of RAID).

Hybrid RAID method

The benefits of this method are:
1. Capacity is 7 drives x usable capacity (N-1 for RAID 5 parity)
2. In this scenario I have used a 100gb SSD, but in reality a 32gb would be sufficient to load an OS, and would result in greater overall capacity in the server
3. The server will boot quickly as it is reading from an SSD during the boot process, and booting an OS mostly consists of reading from the boot disk (hence the fast SSD)
4. The unused disk space on each disk can be utilized in another array if required – some users may wish to do this, while others will opt for simplicity in design and just not utilize the unused disk space on each 3tb drive.
5. In the above scenario, a 500gb RAID5 disk could be created across the 6 unused disk segments.

Overall benefits of going from traditional to hybrid method:
1. Greater overall capacity in server
2. Faster boot time

Usage scenarios – data center or anywhere that overall capacity is of paramount importance.

Option 2 – Workstation

The workstation user of today wants a fast booting, fast application-loading system without the hassle of having to rebuild a server if a drive fails.

The advantages of this system are:
1. All reads are directed from the SSD. Therefore read speed of the workstation is dramatically improved over that of a standard single hard drive (or even a mirror consisting of two x hdd).
2. Write speed is the same as a standard workstation, but due to the cache on the controller card will be faster than a standard hdd connected to an on-board motherboard disk controller.
3. If either drive fails, the user does not lose data – they simply replace the failed drive and the array will rebuild onto the replacement drive, putting the system back the way it was before the drive failure.
4. Regarding the 200gb capacity unused on the 500gb hdd – it is not recommended to use this additional space (even though it is possible to do so) due to the fact that this data is not redundant, and failure of the hdd would result in data loss.

Option 3 – Small business server

Small business does not mean slow business. Many small business servers require at least some high-performance storage component to handle accounting software, industry-specific small database and even mail servers.
Along with that performance there is always the requirement for capacity. Even small business can easily generate several TB of data in the form of documents, photos, video etc.

The advantages of this system are:
1. The 160gb disk is big enough to create a boot volume for Windows (etc), and leave enough space for a 100gb volume for database function.
2. Because the 160gb hybrid mirror reads directly from the SSD, both OS boot time and database function will be greatly improved over a standard mirror of 2 x hdd.
3. 160gb is lost from the capacity of the data mirror, however with 3tb drives that is a small percentage of capacity loss and does not significantly impact on the usable capacity of the server.

Option 3 – Small business server – extension of the theme

Since the previous configuration requires at least a 4-port RAID card (eg 6405E), the 4th port is unused in the above configuration. It is a simple matter to extend the server to utilize all 4 ports. This would require purchasing a second 160gb SSD to mirror to the unused space from the previous example.

Why?
It might seem far more sensible to mirror the two SSD drives, then mirror the 2 x hdd drives in what would seem a far more conventional server. However, there is good reason to do the above:
1. The write speed of lower-spec solid state drives is not greatly faster than that of HDD, so there is nothing really to be gained in write speed (minor) by mirroring the two SSD.
2. Since the read speed of the SSD is so quick, putting them in a mirror will not result in much improvement over a single SSD since most reads would be concluded before the logic of the RAID card has a chance to bring the read characteristics of the second drive into play.
3. Since the focus is on read speed, having two hybrid raid arrays now gives 320gb of SSD read speed to the server, while still protecting the data on each SSD in the form of a mirror.
4. The capacity of the server (HDD space) is not sacrificed from the previous example.

Option 3 – Small business server – extension of the theme – high-speed, small capacity

This configuration focuses on performance, not capacity.
By purchasing 2 600gb SSD the user can get 1.2tb capacity from these drives. Instead of the traditional method of mirroring the drives and wasting the capacity of the second SSD, by mirroring the SSD to the HDD the user gets the full capacity of the purchased SSD.

This server is for the user who is focused on speed, not capacity, and wishes to utilize all of the SSD capacity.
Note that it would be possible in the above config to add a third SSD to the config (using the 4th port on the RAID card), to give 3 x Hybrid mirrors on a total of 3 SSD and 1 HDD.

Conclusion

As you can see there are plenty of configuration options you can do when you understand Hybrid RAID. While not every one of these may be useful for you, it’s good to keep in mind that there are options out there, and that you can offer the customer differentiation that may just be what he’s looking for.

Ciao
Neil

Posted in Advisor - Neil, Application Environments, General, Platforms, Storage Applications, Storage Interconnects & RAID, Storage Management | Leave a comment

Computex …

Spent a week in Taiwan at Computex – it’s an amazing combination of technology providers and customers from across the world.

PMC had two main focus at this event – one was to work with Chenbro in promoting their new jbod designs and the other was working with EchoStreams in showing their amazing high-speed array technology using our RAID cards.

Chenbro were great – they gave me two servers and 60+ hard drives to work with. It’s not often I get to play with such large numbers of devices and make 220gb+ arrays – my budget doesn’t normally extend to such large amounts of equipment. It’s pretty interesting to look at how dense drive arrays are becoming – 60 drives in a 4U chassis is quite an engineering achievement and the Chenbro equipment worked perfectly all week.

The EchoStreams gear is mind-blowing in performance – check out their website at www.echostreams.com.

Now if I can just get some of this gear into my lab :-)

Ciao
Neil

Posted in Advisor - Neil, Application Environments, General, Platforms, Storage Applications, Storage Interconnects & RAID, Storage Management | Leave a comment

Is it compatible?

Does this work with this? That’s a question that Adaptec often receives from customers who are trying to purchase hard drives to suit a customer application. Very often those drives are new to the customer (hard drive vendors change models as well as changing model names without changing drives faster than we do), so customers get very confused, wary and downright worried about plugging things together.

So what can you do about it?

For starters, look at our compatibility reports. They are very well hidden on our website (something we are trying to fix right now), but they are a good start.

www.adaptec.com/compatibility

(It’s easier to type in the link than try and find it through navigating the website.)

This is a list of what we have tested that works. That’s not necessarily to say that it’s a list of everything we have tested, just what works. Take serious note of the firmware revisions of the devices (especially hard drives) as they play a big part in the reliability and compatibility of products.

There are various notes around the pages (some that I missed reading myself), such as the fact that we don’t support “SED” (self-encrypting drives), but the important one to see is that if a product is of the same family, but not on the list, then we are comfortable that it will work. So if you see a 1TB drive on the list, and you are purchasing a 2TB model of the same family from the same vendor, then it will be OK.

If in doubt call Adaptec, but start with the compatibility lists. Are they completely up to date all the time? No, that would be impossible. Are there going to be products you want to purchase that are not on there? Of course, Murphy was an optimist after all.

But don’t (a) blindly purchase products without doing some research and (b) don’t not use a card because you “think” it might not work. Check first, then call second if you have to, but work on the principle that a very large percentage of products in the market work with just about every vendor’s cards so compatibility should be the exception, not the major sticking point in a system build decision process.

Ciao
Neil

Posted in Advisor - Neil, Application Environments, General, Platforms, Storage Applications, Storage Interconnects & RAID, Storage Management | Leave a comment

Oh my goodness, my backplane has an LSI expander! …

So what?

Adaptec cards are tested to work with a large number of devices. While we do our best to hide our compatibility reports (dig at Marketing but I know that is being fixed) … these reports show that we test our cards with a wide range of motherboards, backplanes and drives.

So go and have a dig in the “Resources” section of our website and see if you can find the compatibility reports for that card you think you might be buying tomorrow!

Ciao
N

Posted in Advisor - Neil, Application Environments, General, Platforms, Storage Applications, Storage Interconnects & RAID, Storage Management | Leave a comment

Power management – are you using it?

Adaptec controllers have, for quite some time now, had the ability to power manage the hard drives attached to the controller. It’s a free feature that watches the data throughput to the drives, and can be set to spin the drives down when they are not used.

So … who is using it?

It’s one of those features that we have on our card that people say they need, but I wonder exactly how many of our customers actually turn this feature on?

Do you really need it or use it? Send me some feedback.

Ciao
Neil

Posted in Advisor - Neil, Application Environments, General, Platforms, Storage Applications, Storage Interconnects & RAID, Storage Management | Leave a comment

Banging on about that card again …

I know I harp about this card, but only because I think it deserves more of a rap than it normally gets … so please read on …

Are you using SSD Drives?

It has become very common for system integrators to use SSD drives instead of 10K or 15K SAS drives when developing high-speed (eg database) servers. When configured in RAID 10 for maximum redundancy and performance, this makes amazingly fast server storage.

Very often you will only be using 4, 6 or 8 drives in this configuration, which sends you on the search for the “right” controller to suit your customer needs (both in performance and budget).

So what is the right controller in this environment?

71605E

Initially you may consider that this is:

(a)    an entry-level card with reduced RAM (therefore not suitable for your needs); or

(b)   a 16-port controller when you only need 4 or 8 ports; or

(c)    that this card does not support cache protection which you wish to promote to your customer for maximum protection and performance; and finally

(d)   you are concerned whether this card’s on-board processor is fast enough to handle the IOPs that the 4 or more SSD are capable of sustaining.

Those are all valid and essential considerations when developing a performance server, but please read the following item which will help you determine that this is in fact the right card for your requirements …

  1. RAID 10 does not require a great deal of memory (very little in fact) for its RAID function, so the 256Mb on the card is more than sufficient – the RAID 10 will perform just as fast on this card as on a card with 1Gb memory.
  2. The fact that this card has 16 ports is just a bonus to you. If you are only using 4 then it acts exactly as a 4-port card. The main consideration here is the price. Please look at your supplier price list to see why the 71605E is in fact a good value card in this respect.
  3. This card does not support cache protection. That is true, but cache protection only protects the data writes sitting in the cache on the RAID card – and the critical factor here is that when using SSD drives only, the controller will prompt you to turn off the cache – SSDs work better when we don’t cache the writes to the array. So the fact that you can’t protect the cache doesn’t matter because there are no writes in the cache (because it’s turned off for the array).
  4. The processor on the 71605E is exactly the same as the processor on the rest of the Adaptec 7 Series cards, and can handle IOPs in excess of 450,000 – which is as fast as any controller on the market today. This is in fact the fastest 6Gb processor on the market today.

So the next time you have a configuration requiring all SSD in a RAID 0, 1 or 10, consider the 71605E – it is much, much more than an entry-level card in this environment.

Posted in Advisor - Neil, Application Environments, General, Platforms, Storage Applications, Storage Interconnects & RAID, Storage Management | Leave a comment