The IT Business

The world celebrates the 11th SysAdminDay today. To all the users out there: This is a nice opportunity to thank your admin staff for doing migrations at night to keep your business running at day, for making seemingly impossible requirements possible, for keeping the systems up and running, for answering your questions, that would be able to answer yourself with google in the time you need to pick up the phone

And keep in mind, your admin would do everything for you to restore the operations of your datacenter:

Is this a new trend in IT sales? Companies offering AMEX gift cards for attending a sales presentation. Got the second mail of this kind. Greenbytes offered $50 for listening to them, a few minutes ago i've got a similar mail from a company called Exagrid offering $100 for an half hour web presentation.

Somehow i'm not sure what i should think about that: Should i translate this "We are so sure that you will buy the product that we pay you to take a sales call!" or "We are so desperate that we need to pay people just to take a sales call...".

Perhaps you could make a living out of professional sales attending ... given that such examples are representing a trend ... or auction an hour a day on eBay in half-hour increments for such paid sales calls

With this article i want to learn a thing or two and i want to start a discussion.

Whenever you express, that a mainframe hasn't no advantages over a high-end Unix system, the first comment or reply to a comment is "Oh, but IBM has Parallel Sysplex and delivers availability xyz can only dream of". Often the discussion ends there, because few people know, what this "Parallel Sysplex" is. I had to read into a lot of documentation to learn more about the Parallel Sysplex.

I've learned a few things in that: A sysplex doesn't seem to be much different from something like we simply call a cluster in Unix. A parallel sysplex is an extension of this and enables the system to act as one large system. However you can't consider it as a tool that is able to distribute arbitrary tasks on sub-process level onto several systems.

As far as i understand the concept of the Parallel Sysplex, it's not the way that any and every software is enabled to be a part of a parallel sysplex as it has to use certain facilities that make a parallel sysplex out of a sysplex. Your application have to use a thing called the coupling facility. And parallel sysplex exactly contributes this facility to the applications running on the mainframe.

Parallel Sysplex doesn't help you to make your zLinux more available. You have to work with the Linux cluster-frameworks to do so, albeit there seems to be some workarounds

I understand that the parallel sysplex makes DB2 on z/OS more scalable and available, but where is the difference to Oracle RAC on Solaris on a 2 big node cluster? I understand that a parallel sysplex makes CICS more available and scaleable, but where is that different for example from using Oracle Tuxedo in a highly available configuration with clustering support? I understand that that parallel sysplex can make storage more available, but how is that different from the Availability Suite with it's Remote mirror facility, from SamFS/QFS, from the global filesystem in Sun Cluster.

I know, all these things working differently, but that's not the point. I want to know, why some people think "Oracle can just dream of". What's so special about that ... the parallel sysplex is an ... well ... interesting architecture, however i fail to see any reason why it's special that everyone talking about mainframes is just pointing to it.

Perhaps it's just the way that mainframe people doesn't understand the open systems world to the same extend as open systems people understand the mainframe world: Barely to an extend at all. Additionally i have the impression that many services on a mainframe are so ubiquious in that world that they consider properties of this software components as properties of the mainframes. Another hypothesis is, that most mainframe guys doesn't know what's possible on Unix (real Unix, not the pengiun variant) and thus consider their technologies as superior. I talked to a mainframe guy on a conference some time ago and i had to explain that SunCluster has gone great lengths since Sun Cluster 2.

We talk about synchronous storage replication in Solaris for quite a time now (2000), service restarting isn't a problem since SMF. Instruction Retry on hardware basis is available since the introduction of the M-Series, the same for memory mirroring. Hot swap of I/O, memory and CPUS is possible since last century.

May be i just don't get get, but may you are able to help me: I want to postulate "There is nothing in a Parallel Sysplex that a equivalent configuation on the basis of an M9000 isn't capable to do". But i'm a open systems guy from the core, so i ask the mainframe aware people to explaon me, why Parallel Sysplex is considered as superior to Unix-HA clustering when used with applications developed for high availability. Please don't answer with "You have no idea, ParSys is so superior" ... i want some real reasons.

There is one point in the media reporting that is really upsetting me: Some news outlets were touting the hybrid mainframe as the development that will save this technology for decades. IBM gets an overly positive press for placing a rack besides the mainframe.

Sun was talking about the concept of workload computing for years now: Use x86, UltraSPARC T* and SPARC64 where it fits best and use Solaris as the overarching umbrella. The same OS on all components.

However nobody was touting this as a major step forward ... Just as a reason to bash Sun, because they interpreted is as uncertainity about their technology. There were even people seeing this as an excuse for the single thread performance of T1/T2.

They didn't looked at it as the basic point behind it: That one size doesn't fit all and one proc doesn't fit all needs, and a proc can't fulfill all needs.

Just got the news, that Oracle Solaris Premier (24x7 Support) Support is available on HP servers via HP again. You have to order it via HP Software License Management Services. For details ask your friendly HP sales rep. However:there is (and was) a way to buy Support for Solaris on HP via Oracle, about that way ask you friendly Oracle sales rep.

When i've read first about this hybrid mainframe concept, i imagined some weird concept to couple mainframes and x86 servers to work in conjunction and the technologist in me thought "Cool ... how did they done that?". And in the press the concept of the hybrid mainframe found some resonation.

The reality however is much more simpler ... more in the direction of the Sun Fire USBRDT 5240 Uniboard

The nice thing about IBM is their RedPaper program, because you will find information somewhat unhampered by marketing without the need to wade through many manuals. So i've read through the paper IBM zEnterprise BladeCenter Extension Model 001.

What is the hybrid mainframe? The hardware? Well a mainframe with a rack of BladeCenters. The highspeed network in between the mainframe? 10 Gigabit Ethernet.

Okay, nice ... but not exactly rocket science. At the moment there is just one solution that supports this, an accelerator for analytical work on mainframe DB2. This concept remembers me of ideas like the SAP BI accelerator, the storage cells in Exadata and concepts like that that offloads certain tasks to intelligent nodes. Furthermore: It looks like that the integration is primarily a job (and thus a property) of the database, not of the mainframe. At the moment this hybrid mainframe stuff it's really just some way to accelerate Business Intelligence queries.

Sorry ... but this mainframe is as hybrid as an Sun M9000 with a rack of x86 at the side, both using Solaris, would be. Putting a grid engine in front of it would make it even more hybrid ... as you could submit arbitrary jobs on your submit hosts and the grid engine would execute the job on the matching architecture

I assume it would be much cheaper just to use standard x86 servers, but that would destroy the story of mainframes, as it would be more obvious that they are needing more help from cheap x86 servers.

As a technology guy, i'm a little bit disappointed. Boooooring!

PS: However the integration of the USBRDT was higher ... it got at least power and cooling from the chassis making the E25k the most expensive blade chassis of the world

While trying to bring down my unread articles counter to zero, i've read an press release of SGI about their UV1000 SPECjbb2005 benchmark. And indeed: Some of the numbers are impressive.

However, there is a second story in this numbers. It's the story of the long arm of Mr. Amdahl - again. And it it underlines my opinion, that we need mandatory SPECjbb2005 results with 1 JVM at 64 bits I have to thank SGI for giving me some data points to check my hypothesis.

IBM opened up a beta programm for AIX 7. The "AIX 5.2 Workload Partitions for AIX" feature looks pretty much like Solaris Containers for Solaris 9 and 8. Giving a Zone control over a device is an old hat under Solaris 10. And i don't know if i would brag about LACP based link-aggregation in 2010 ....

I assume i wasn't able to clarify my point versus the "multitude of JVM" SPECjbb2005 results. Of course there are many applications out there where the scaling behaviour mandates many small JVM instances. And there is nothing to win but performance to lose to use a 64-bit for an application that just uses one gigabyte of memory.

However that isn't the class of problem you try to solve with those large machines that are used in those benchmarks SPECjbb2005 benchmarks.

What's the sense of purchasing a p750 with 16 cores just to divide it in 16 JVM. What's the point of buying a p595, a Altix UV, a M9000 just to slice and dice it. When you look at many "multitude of JVMs" results, you even see commands to bind them to a CPU disallowing the process to migrate around on the system, making a group of 1 proc servers out of this large and mighty machines. Such a configuration is only justifiable by the much better RAS of such system.

In such a case it can better to purchase some M3000 instead of M9000. Many people forget that the capability to scale to a large number of CPUs doesn't come for cheap ... hardware scalability costs performance for example by introducing a slightly longer latency by a larger crossbar or multiple stages of crossbars. When you system doesn't need it, don't use it. But when your system needs large CPU counts or large memory heaps, SPECjbb2005 results may be not indicative for the performance you get, because the vendors factored out something important from the result.

Wouldn't it be more obvious just to purchase several smaller system when my application just doesn't use (and can't use because of the 32-bitness) the main advantage of such systems like providing a really large single memory area. Just to reiterate it: When you want to buy a cluster, you should buy one. This multitude of JVM configuration transform SPECjbb in a easily clusterable application. It's misguiding about the scaling behaviour of the system. But i'm repeating my self ... i would really like to see the dismissal of "multitude of JVM" results or at least a mandatory "1JVM/64 bit" result as a SPECjbb2005_base result.

There are some changes to the article i've ranted about. But at first i should note, that i was right about my opinion that Ronan Miles didn't told silicon.com, that Oracle has canceled x86 servers. Stefan Parvu simply asked Mr. Miles about this quote. The answer of Ronan as reported by Stefan in the solarisx86 newsgroup:

That is a complete mis-quote, never said anything of the sort about removal of
technology. Sadly, journo's look for soundbytes.

Thanks for the link, I will follow up.

It looks like this "follow-up" was successful. The article changed from

President of the UK Oracle User Group Ronan Miles told silicon.com that Oracle has achieved much more on the hardware side of things than many expected through the rationalisation of the Sun business, including the removal of technology such as X86 servers.

to

President of the UK Oracle User Group Ronan Miles told silicon.com that Oracle has achieved much more on the hardware side of things than many expected through the rationalisation of the Sun business.

Of course there is no notice, that the article has changed. And of course the comment about killing x86 (now only attributed to Mrs Breen) is total nonsense.

According to the Coraid website, NetApp has send a "nice" letter to Coraid, a vendor of an ZFS based storage appliance. NetApp should stop to bully around small vendors, get back to innovation and sort that thing out with someone in the same or a larger weight class, for example with Oracle. I can just explain this behaviour by increasing pressure in the market by ZFS.

PS: At the end the lawsuit is still valid, as far as i know most of the relevant patents has been invalidated ....

A few weeks ago i was very upset about an article in silicon.de totally misinterpreting the matching support policy. Yesterday i put a link into del.icio.us to an article in silicon.com titled "How Oracle has made Sun rise again" to save if for an article and as a "no-comment statement". This article provides a gem like this one:

President of the UK Oracle User Group Ronan Miles told silicon.com that Oracle has achieved much more on the hardware side of things than many expected through the rationalisation of the Sun business, including the removal of technology such as X86 servers.

Suuuuure .... it's just the way that we announced several new x86 servers a few days ago and that Larry is standing in front of a rack full of x86 servers in the picture in the same article. I can't believe that the lead of the Oracle User Group is that uninformed and somehow i tend to think that the part after the comma wasn't said this way. And even if Mr. Miles said this way, i'm expecting some checks and research from a journalist instead of just writing down everything someone is saying. That's the job of a journalist.

With media outlets like this you don't need enemies ...

In der ZEIT ist ein Artikel zu finden, in dem SAP und Oracle miteinander verglichen werden. Ich finde zwar, das Oracle darin etwas zu negativ und SAP einen Hauch zu positiv gesehen wird, aber dennoch durchaus lesbar: "Mit den Waffen des Feindes"

Okay ... i really thought that enforcing a certain brand of cables was a thing of the past. Apparently not as reported by @randybias on his Twitter timeline:

This is really a common problem as shown here or here. Especially the later one is showing the pain you could experience when you use those 10GbE direct attach cables . You have to think about this DAC like two hardwired transceivers). As both ends contain the transceiver logic it's easy to implement an ID rom in that cable and check it before bringing up the link.

I should note, that there is a standard for the interface that connects those DACs to the switches/NICs. It's called SFF 8431. But now lets look into reality: The Intel card of the user in the second links doesn't support the DAC cables from HP. HP seems to just support HP cables, not allowing anyone to use other cables by software. Switches from HP are equally rigid, they just support HP branded DACs and nothing else. Cisco Nexus switches seems to have at least a mechanism to allow unsupported DACs to work to get HP cables running . Randy wasn't able to use Arista DACs with HP Flex 10.

Welcome in the world of cable-locks, where the intersection of supported with one systems and supported with the other system may be empty I thought, standards were made for a reason.

Dave Lutz wrote a great article about partition alignment. This is an important technique to optimize your storage.

If you create and access logical disks (aka LUNs) from your Sun Unified Storage appliance, whether over iSCSI or FibreChannel, you should be aware that client side partition alignment can have a big impact on performance. This is a generic issue that applies to any virtual disk interface, not just Unified Storage, and relates to how client generated virtual disk I/O maps to actual I/O in the appliance. The good news is that it can be quite easy to properly align partitions.

Aligning application block sizes to file system block sizes is a common technique, but you shouldn't stop there. When you are not aligning the all block sizes to the native structure of your storage system you end up with reading or writing additional data because of the misalignment, thus reducing the performance of your storage unnecessarily.

The reason is obvious. You can't read or write bytes from or to a disk, you can just read and write blocks. It's called block device for a reason. When a file system block aligns to the native storage block you have to read just this block. If it's not aligned, you have to read the block with the first half and the block with the second half, thus you've doubled the reads.

It's really easy to prevent such an behaviour, and Dave's article explains the necessary steps to do so. Really worth a read.