SSDs – It’s all about moving the bottleneck January 18, 2009

I decided to put my propeller back on “spin” for a quick post.  Ignore that grinding sound, that’s just the propeller transmission grinding from lack of use.  Lately I’ve been posting about personal stuff, for this one I’ll get back to biz.

 

So – SSDs, or “solid state drives” – what’s the big deal?

 

Well, if you’ve not spent much time learning about these, read the following.  If you already have, move on to the next paragraph or send me corrections to the following one.

 

Solid State Drives are basically hard drives, but with no moving parts.  Instead of a rotating platter and heads that read the strings of zeros and ones stored on the magnetic coating of the platter, an SSD uses some sort of memory technology and has no moving parts.  Why is this a good idea?  Simple – speed.  Believe it or not, when you have a disk drive with platters that rotate at 15K RPM and drive heads that move a maximum of an inch, if the drive is reading the data from the inside drive track (the one closest to the spindle) it takes a long time to get the head to the outside track of the platter.  Sure, those itty bitty heads move fast and sure that drive platter is spinning fast as well, but in computer terms – it takes an eon to get that head to move to read that outside track.

 

This is one of the reasons drive vendors add cache to their drives.  The basic idea is simple – if your PC is wanting to read a given sector off a hard drive, there is a good probability that it wants to read the next sector as well.  To make that second read as fast as possible, the drive doesn’t wait for the system to ask for the data, it just goes on and reads it anyway, then stores it in the cache on the drive.

 

Sounds good, and it is.  This works great, has for years.  But, there’s a catch (surprised?).

 

Reading the next sector and storing it in cache sure does make it faster to read, but what if your PC didn’t want that next sector?  What if it really wanted a sector on a different part of the drive?  Well, that stored read is not used and the (microseconds) it took to get that data are wasted.  That’s called a “cache miss.”  Cache misses occur when PC, in particular a server or storage array, accesses drives randomly (true random access will cause a 100% cache miss, every time).  When on earth would access be that random?  Databases……….

 

In the case of a database (and, no, it does not matter which one – MySQL, SQL, Oracle, etc) the data in them is often access randomly.  The database doesn’t know what record will be asked for next, so neither it nor the drive can accurately predict what “next” sector to read to put into cache.

 

Dang, now what?  Ahhhhh…….enter SSDs.

 

SSDs, like their name indicates, are solid state devices – they have no moving parts.  No spinning platters, no moving heads, no servo motors.  Why is this a good thing?

 

Remember – when you want to access data randomly from a conventional (rotating) drive, you may have to wait for that head to move from the inside track to the outside – with an SSD there is no head.  No waiting.  Random access is just as fast as sequential access.  How cool is that?

 

Turns out, very cool indeed.  Whether or not you believe the published benchmark numbers, the fact that SSDs are faster than even the fastest 15K ROM FC drives can’t be argued.  An SSD will generally outperform a large number of conventional drives, no question.

 

So – why not just load up every storage array with SSDs and have the fastest storage array on earth?  Turns out, it is all about moving the bottleneck.  Once you understand where the bottlenecks are, you can make some more informed choices about moving them, and designing in balance in your systems.

 

Suppose you have a storage array with, say, slots for a dozen hard drives.  Suppose you are currently running in that storage array a dozen 15K RPM SAS drives in, say, a single RAID 5 array (the details of whether or not this is optimal aren’t relevant for this discussion).  Suppose that array is managed by a PCIe RAID controller in a x4 slot.  Suppose that array is connected to the rest of the world via a single 4 gig FC port.

 

So – where’s the bottleneck in random I/O?  Easy.  I can tell you w/o even benchmarking the system.  The bottleneck is the drives.  How do you know?  Simple – rotating drives, unless used in massive quantities, are always the bottleneck w/random I/O.  It isn’t the controller, it isn’t the controller’s bus, and it isn’t the FC connection.  It is the drives.

 

Now, what happens when you add SSDs to that same array and look for the bottleneck?  Well, it gets interesting, that’s what.

 

Now you have to start asking some questions:

 

Q: Can the processor on the RAID card keep pace with parity bit calculation if I use SSDs?  That is, will the processor on the card be overwhelmed?

Q: If not, what about the bus the RAID card uses?  Is a PCIe gen 1 x4 enough?

Q: If it is, what about that 4 gig FC connection?

 

See?  I told you it gets interesting!

 

I won’t work through all the numbers, but as you can see SSDs can be a real game changer.  Their incredibly high I/O and throughput speeds force us to think a bit more about designing storage systems.  They force us to ask – where’s the bottleneck?

 

Next post – some interesting s/w that exploits the speed of SSDs, but doesn’t require that you ditch all your rotating disks.

The (job) hunt continues… January 14, 2009

First and foremost – thank you to all the people who have reached out and offered to help in my job search.  I sincerely appreciate it.  As many of you already know, looking for work (particularly in the tech sector) is tough right now.  I’m privileged to have so many folks sending me tips on jobs, offering to be references, forwarding my resume to their co-workers.  It is, frankly, making these tough times a little easier.  Thank you.

I have been making some progress getting support for former Gateway and MPC customers lined up.  I know you may be tired of hearing this, but please be patient (remember – I’m looking for a job as well).  I have heard back from Gateway’s manufacturing partner and have a good lead on one of their partners who may help (I will contact them tomorrow) and am also making progress with Intel for the former MPC customers.

Speaking of Intel – another public “thank you” is due to them.  I can’t say too much about why, but if you are one of the good folks that I now from Intel who has been e-mailing me recently, you know why.  As more develops, I’ll post more details, but it is really too early now.

Last – a big thanks to my former co-workers.  You are all a great bunch of people.  Thank you for trusting me to contact your customers directly, thank you also for your support.

Is there life after MPC/Gateway Pro, for our customers? January 9, 2009

As per my first post, the fun is now over.  I got the call – MPC was shutting down, my services were no longer required.  Bummer. 

I was, and still am, quite unhappy in a completely selfish way – my family has to eat, the bank still wants their mortgage payment, still have to pay car insurance, and now need to reserver even more for meds.  But that’s all me – who else has this affected?

By far and away the largest group of affected people is the (former) MPC/Gateway Pro customer base.  How might they get support?  What should they do if a key part in their Gateway Pro server – which happens to be running their Exchange – fails?  What is their first step if a part in their MPC server fails?  Where do they even start?

Keep in mind – some of these customers really stuck their neck out buying from either company.  They could have gone with the “big boys” (you know who I mean), but didn’t.  At (probably) some risk to their career, they picked us.  Now?  Lucky them.

So what’s to be done about it?

Well, unfortunately, it seems some folks have what I think is the _wrong_ idea.  Some of those same folks are going to get angry reading this.  If you do, sorry – but it’s a blog, nobody is forcing you to read it.

I feel that some are trying to exploit what must be a very difficult situation for these (former) customers (I hate having to add that, but want to be clear that since the company has folded, they aren’t really customers now) by offering “private” support contracts, or keeping their customer contact information to themselves, for fear that someone else might do the same.

From where I sit, our (former) customers have been through enough and “when I have time” support isn’t going to help them much.  I’d guess (and it is only a guess) that some former employees have very tight relationships with customers with whom they’ve worked and really do want to try and help.  Others, I’m not so sure about.

So – putting my money where my mouth is – I’m trying to set up contacts at Gateway Pro and MPC’s former server ODMs to see how/if they might be interested in helping make this bad situation a little less so.  I’m happy to report that progress is being made.  Both Gateway Pro and MPC’s server ODM partners have – at least so far – expressed interest in helping our (former) customers.  For a fee, of course.

But everyone has to eat, right?

I’m not suggesting that we former employees “give away” anything, quite the contrary.  What I’m suggesting is that if you have a mind to help a former customer as a private individual or as a member of a company who offers services (if you are astute enough to have landed a job already – good for you), as you consider what – if anything – you want to charge, also consider that many customers have already paid for service at the time of purchase.  Service they won’t get.  Go easy on them, that’s all.

Spare parts will always have to cost something.  The folks supplying the part didn’t make the margin on the initial sale, so that’s expected.  Shipping charges will also have to factor  in.

All I’m hoping for is that some of you will maybe dispense a little free advice to the folks who helped us be as successful as we were and who put dinner on our tables and helped make our mortgage payments.

Access denied! January 8, 2009

If you read my previous post, or if you have tried to secure health insurance on your own and are not the model of youth and fitness, you have probably been greeted by the above expression after filing your health insurance application:

Access Denied!!!

Yup, got that one today.  I get it.  Health insurance is legalized gambling: you are betting that you will get sick and it will cost a bundle; the insurance company is betting you won’t and they’ll make some money off you.  Nothing wrong w/that, right?

Well………….while it is entirely reasonable to raise premiums to cover the cost of high risk applicants, the complete refusal to provide health insurance isn’t.  As my wife put it last night – “If they don’t offer coverage to those who need it, what’s the point?”  My wife has a way of cutting through the baloney and getting right to it.

So, what’ the next option?  Enter a high risk pool, accept state mandated coverage (looking in to what that provides, but probably not a lot), or select the COBRA coverage offered by my former employer (well, what’s left of them).  Looks like that COBRA policy may be an OK deal after all.

On a brighter note – I’d like to express public gratitude to the folks from Gateway’s former ODM as well as Intel for being interested in trying to help MPC/Gateway Pro former customers.  I have gotten encouraging replies from each company and hope to be able to point our former customers in a clear direction to obtain needed parts and service.  A (serious) public “thank you” to both companies.

And on a still brighter note – my wife is feeling better and my son seems to be on the way.  Thanks to all that contacted me.

A scary feeling last night…very scary… January 4, 2009

If anyone read my first post, you know that – like many other Treasure Valley residents – I am out of work.  Sure that means no income, but worse that that – it means my wife and I have to pay full list retail for doctor visits and prescriptions.  Why is that more scary than no income?  Simple – I’m diabetic and my son is not yet two.

What’s the big deal you ask?  Can’t I just test my blood and take shots?  Sure, that’s the easy part.  Buying the supplies, now that’s a bit tougher.  Consider:

Blood test strips – just under a dollar each, I use about 7 a day.

Insulin – about $400/month

Blood pressure – 2 meds.  Fortunately, one is on Walmart’s $4 list, but the other is about $2/day.

What’s that all add up to? $400 (insulin) + $210 (blood strips) + $4 (one blood pressure med) + $60 (other blood pressure med) = $674/month in meds.  Just for me.

Which brings me to the “scary” part.  My wife and my son are both very ill.  Since last night each has been vomiting regularly.  It is probably something they ate (i didn’t have the cheese-macs, they did, I’m fine).  What thought was in the forefront of my thoughts?  “What can I do to help them feel better?”  “Oh my goodnesss, I wish it was me?”  Nope, those would be easy.

I’m embarassed to admit that my foremost thought was “Please don’t let them have to go to the hospital – that will completely drain the savings we’ve been counting on to get us through until I find work.”  Does that suck or what?

I never took health insurance and drug coverage for granted.  I’ve been diabetic for 30+ years and been on Medicaid before, so I realize the value of health insurance.

What was new this time was the consideration that – should my family need to go to the hospitak – it could put us in a very tough situation as I try and find work.

And that’s what is scary, really scary.

Another Idaho company closes… January 3, 2009

Monday of this week was a sad day.   Another Idaho company closed its doors, this time a little closer to home.

Rewind the clock to 1995.  My wife (then girlfriend) and I arrived in Idaho for me to interview with an up and coming PC manufacturer.  We did some research (remember – this was well before the Internet was pervasive) and arrived in Nampa bright and early one morning.  Initially we wondered what we were in for: tumbleweeds blew across Karcher road (literally) as we sat in front of this tin shed in what seemed the middle of nowhere.  I interviewed, she waited.  A few days letter we got an offer.  A good offer.  We packed up and moved to Nampa (we were from out of the area).  Young and inexperienced, we were elated to start a new life here.  We bought a house, we bought a couple of (used) cars.  We loved the area.  We were on our way.

Fast forward to this past Monday – the fun is over. 

I got a call from my (former) boss around 9 AM telling me that the company is closing and today is the last day of operations (keep in mind – most of us were on vacation).  The top level execs were still there, but the vast majority of the rest were out.

Before anyone thinks I’m crying “poor me” I’m not.  But the reality is that we, like so many others, will probably have to leave the area now.  That’s a shame.  There are enough vacant homes here, enough empty store fronts, enough of an exodus.  Enough………….