Posts by Billl 165 publicly visible posts • joined 15 Jul 2009
Hey Mr. Sunsetter, HP is turning into Dell, and you know it. HP will not be able to compete with Dell. This is why they brought on Leo, 'cuz they need software. Of course, Sun was trying to do the same thing when their market was shrinking... luckily for them they found the SW Nirvana when they were acquired by Oracle.
Unless HP gets bought out or merges with SAP, HP will just be another Dell.
I agree that a feature like this takes time to test, but from my understanding no one really uses this feature on IBM HW, as all of requirements to make it work properly are next to impossible to meet. Correct me if I am wrong... this is what I hear from friends that use Power systems (I do not).
Regarding the cache comment on the number of partitions... bull. Sun nor Oracle make any such recommendation based on cache size. SPARC does not have to have as much cache as Power since it is CMT and can get work done even while a thread is waiting on a memory load. Power will stall all threads on a cache miss, which will happen regardless of how much cache you have. When you slice up a CPU, like IBM does, you WILL have a lot of cache misses as each partition will be doing different things.
One generation and just stating the word "more" does not make a roadmap. IBM still has the worst roadmap of the three (Oracle, Intel, IBM). Of course, as many have said, they do not have as much need to be public with their roadmaps as Intel and Oracle. They've had trouble meeting the timelines of their roadmaps in the past (like Oracle and Intel), so it makes sense to be vague. This way they can say they've met all of their promises.
... and neither do IBM marketing droids... d'oh!
The SPARC64 FUD is getting old. Who cares? SPARC is SPARC. Oracles roadmap is clear that there will be a high end SPARC chip to follow up SPARC64 -- who knows if it'll be SPARC64 or good ol' SPARC. It all runs Solaris with full binary compatibility.
I've always kinda wondered that myself. I think it has something to do with the resellers. The manufacturers don't always know all the details of what and when they sell, or to whom. I think the analyst numbers are heralded or dismissed by the vendors cuz it's a third party validation of what they're doing.
Also, it is well know that many vendors will put a ton of boxes in a truck and call them sold this or that quarter, when in fact they are sold in a different quarter. You can't necessarily trust the vendors numbers, as accounting rules can be manipulated.
I'm sure TPM is well along on another article comparing and contrasting the numbers for IDC and Gartner, but in the mean time check out what Gartner is saying:
http://www.marketwatch.com/story/gartner-says-first-quarter-server-shipments-rise-2011-05-26
Oracle increased Revenues by a staggering 34%. More than all top 5 server vendors.
Just glad to see some real competition in the server market again. Welcome back Oracle/Sun!
"Strange, because I've been known to say exactly that, so are you disagreeing with yourself or just admitting you don't actually read my posts before frothing up?"
SplitBrains comment had little to no froth that I could detect. You, on the other hand, have a bit of spittle on the corner of your mouth Matt. You may say what SB said, but then you go on with your rants and FUD anyway. You can't have it both ways. You're either a fanboi or a thoughtful IT professional.
I cringe whenever Keb comments, as he is unabashedly a fanboi. At least he's honest about it.
A beer, cuz we all have our preferences (mine is all beer in general), and I think Matt needs to chill a bit.
I don't think that's fair. If you were referring to Sun from a year ago, I would agree, but the fact that Oracle seems to have turned around Sun's business in only a years time is nothing short of miraculous. I hate it when it happens, but I have to agree with MB's comment about if Oracles growth is from displacing Fujitsu, or if it is "real" growth...
Wow TPM, your bias knows no bounds. Oracle has been bleeding for so long, and now that they have the second highest growth of all the major server vendors you say they "eked" out some growth? Since when is double digit growth defined as "eked"? You have officially lost all credibility.
"More and more effort seems to go in the direction of not actually producing anything towards a parasitic life form."
I agree that something is wrong in America. Too much time spent punishing success and encouraging/propping up failure (individual/institutional/corporate).
However, I'm not sure how it is parasitic for one company to protect what they have created. If the New York Times or The Times or whatever creates a newspaper, and spends their money paying people to create the articles and take the pictures and then someone just copies that newspaper and just changes the name, but charges less money... that's wrong! What's the difference?
American companies are spending more on R&D per year than most European countries take in taxes. Protecting property rights is key to companies continuing to invest in property (intellectual as well as physical).
There is clear reason for Oracle to demonstrate that Google and ASF knowingly acted inappropriately. I doubt that they are looking for any conversation stating that they need to stay compliant. They are looking for actual collusion. Such as:
"That bit of code not compliant."
"I know, but no one will notice."
The US courts decided that SCO did not have standing, regardless of whether the code was copied or not. At the time, Novell probably could have sued, as they actually had the Patent/Copyrights that SCO was claiming to have.
This is completely different than the case here with Oracle's ownership of Java. No one is claiming that Oracle does not have standing.
ummm... Solaris Zones and BSD jails maybe? Even chroot jails on *nix is reasonably effective.
There -- solved. What else you got?
VMware/Xen do not only provide better utilization, they also provide a way to consolidate heterogeneous environments. If Homogeneous consolidation was the only problem, then the above mentioned solutions would be fine and we would not need VMware/Xen.
"The Chinese and the French know this..."
I tend to agree that the Chinese government and the French government have more in common than the French like to admit, but I am surprised to see some like yourself admit as much.
France makes me sad, as they were once a great protector of individual liberties, as was the US -- at one time.
Intel burned themselves. They made such an effort to make Itanium so complex that no one could copy it like X86 that they completely ignored the need to be compatible with x86. They wanted to leave the x86 behind... and leave the copiers behind. That was Intels fault and AMD made them rue the decision.
If AMD had not come up with 64-bit extensions, then Intel would still be pushing Itanium as hard as they were before. We all have AMD to thank for Xeon.
Funny, asdf, you make Matt sound reasonable... You throw out unsubstantiated comments at every mention of SPARC. At least Matt acknowledged a personal animosity toward Sun Sales Reps. What's your problem? Panties a bit tight?
With the money that Oracle is throwing at SPARC, it sounds more and more likely that SPARC will have a comeback, starting with the T4. Fujitsu apparently has bought into it. Better than Intel and HP making their customers migrate once again to another architecture...
I hate agreeing with the super fudster when attacking the lame fudster, but this "black and white" commitment is pretty weak, for all the reasons that Allison points out. So, Intel and HP, what does come after Kittson? Seeing how Kittson will be released ~2014-2015, it is not hard to expect that it will be supported for 5-6 years at least, which puts us into the 10yr range that Intel and HP are espousing. Seeing how nothing will be running on it, how can they seriously expect us to buy what they're selling?
Settle down Francis. As TPM said, IBM does not provide PUBLIC roadmaps (not very good ones at least). If TPM has seen the NDA Roadmap, then he would have to sign the NDA and could not publish it. I am assuming that since you've seen the "Executive presentations" then you have signed a NDA and should not be publishing your statements such as you are.
Now if someone were to let the IBM NDA Roadmap leak, then I believe that TPM could link to it...
Wow, envious ever? IDC has stated the numbers are actually better than what Oracle is saying and they expect over 4Billion by 2013 in Exadata alone.
When will Oracles "competition" stop FUDing Exadata and start actually making a competitive product? HP just adds a bunch of SSDs and IBM buys a low end solution that doesn't even use their "Flagship" DB2...
Show up or shut up Allison. Your FUD is very transparent and unflattering.
"SPARC is Dead on arrival" -- what does that even mean?
"...why do you assume this is an Itanic Killer?"
Because:
1. Intel states that this chip can run any load in the world -- which I conclude includes Itanic
loads.
2. Intel has moved high-end RAS features into this chip, which makes you wonder why you
really need Itanic?
3. QPI is common between the chips, removing any kind of value Itanic may of had.
4. Because Oracle started the rumor that Intel planned to kill Itanic and this new chip just feeds
into that storyline.
How again is Intel not trying to kill Itanium? HP-UX, if not in the process of being ported, will have to be, or will they go the route of running HP-UX on an interpreter? That would be in keeping with HP-UX's and Intel's promise of supporting HP-UX for many years to come and still prove Oracle right that Itanic is dead.
Will Oracle then support HP-UX on Xeon?
I'll forgive you for not paying attention ;). Sun screwed the pooch big time on US-V and ROCK, but this is now Oracle and I'm giving them the benefit of the doubt -- at least with regard to SPARC roadmaps. Now for IBM's misses:
2003 - Power6 was originally scheduled for 2006
2004 - Power6 now 2006-2007
2005 - Power7 first discussed and scheduled for 2008-2009
and Power6+ in 2008-2009 (different roadmaps)
2006 - Power6 now 2007-2009 and Power7 2010-2011.
2007 - First Power6 system and just the 570.
2008 - Full rollout of Power6
2009 - Power6+ reannounced... Lots of confusion with Power6+...
2010 - Power7 finally released?
They've been late with every version. I can point you at the docs that show these dates... I still have them all...
Really? Poulson (with little detail) and "Future Kittson Processor" counts as a road map to you? Their public road map goes out to 2011 and "future". Now, I am not a big fan of how Oracle keeps things close to the breast, but they have been the most open about future SPARC road maps -- you'd have to be blind and naive to think otherwise, or just biased.
Hell, Oracle even provided TPM, a long time Oracle/Sun protagonist with access to the roadmap:
http://regmedia.co.uk/2010/12/03/oracle_sparc_roadmap.jpg
You may reply back "HP has a better NDA roadmap!" As you know, vendors generally feel that they can change a NDA road map as they see fit. A public road map, while still not a contractual obligation, has more standing as they must publicly stand by it... IBM is notorious for pushing out release dates on the NDA road maps and then creating a public road map when the date is actually set in stone (TPM falls for this all the time setting the impression that IBM is better at meeting their time lines...).
Sun/Oracle will need to meet a hell of a lot of road map goals to make up for ROCK, but Itanium has been a total joke when it comes to meeting time lines.
Whether this is true or not, I don't think it matters. The roadmap shows a SPARC CPU taking over performance-wise (throughput AND single thread perf). This whole SPARC64 is dead thing is just a red herring. Who cares? SPARC is an open technology and Solaris will run on SPARC regardless of who makes it -- no recompiling! You don't get the same thing with Itanic nor Power -- they are truly propriety. Sun has used third party SPARC way before SPARC64 came around and Oracle may well use it in the future. Get some new FUD!
The EU is as impotent as the UN. They backed off of the Oracle/Sun merger, after huffing and puffing for a while, and Larry got everything he wanted with little or no concessions.
SAP and HP should hurry up and get on with it -- merge already! The US Justice Department will not have a problem, but I'm sure that the EU would be very upset to lose the largest European tech company to a US company -- or would it be a merger of equals like the HP/Compaq merger (LOL)?
Pretty funny... TPM may be biased toward Power, but he has not been unfair to Itanic. Oracle has the most detailed roadmap out there. Power and Intel just show names... SPARC shows cores, estimated perf, and sockets supported. You can attack IBM on this fact, but Oracle has been very open.
It was just a joke. Yes, I know you guys on opposite sides of the discussion when it comes to Itanic and Power, but your tired fud when it comes to SPARC is overly repetitive.
The fact is, Sun stated that they designed the "beefy" cores 5 years ago, and are just now getting them out to market. That does not sound like a change in direction, it sounds like clear planning. Increase the thread count, then make those same threads faster once the technology can handle it.
Not a change, just a natural progression.
You seem to make a valid point, but miss the point in the end. The point is that the non-CMT procs, such as SPARC64 VII+, need more cache because the CPU is stalled while waiting for memory. When waiting for memory (which is slow relatively) CMT just runs another thread that has it's data already, and the thread that is waiting for memory can wait as it normally would - there is no cache-miss penalty. SPARC64 (and other non-CMT CPU's) if the CPU has a cache miss (likely), then the whole CPU must stall and nothing gets done until the memory gets there. The instruction set has nothing to do with it (it's not the cores as much as it is the threads).
Very simple really. They could add more cache to a CMT processor, but it really would not make much difference (theoretically of course). If you did increase the speed of the CMT processor, then it is possible (I guess) that more cache would help a small bit, but without doing it, who knows?
Jesper, do you spend a lot of time at the pub with your other marketing buddy, Matt, coming up with nonsense fud? If Increasing the clock on SPARC means Sun/Oracle was wrong, then Intel and IBM going to more threads and cores means they were wrong????? Take a basic logic course. This is the logical progression of chip design. Go to a new chip design. Figure out how to get more cores/threads on a processor, then figure out how to make it more general purpose by increasing the per thread perf....
End to end data integrity. Look it up Matt. No other general purpose filesystem does this (even WAFL). To be fair WAFL does do a basic WAFL, but not to the same extent that ZFS does. You may write the data to disk, and it may be mirrored, but did you know that it was written wrong, and that data was written wrong on both mirrors? ZFS knows it. WAFL sometimes knows it. No other filesystem knows it.
Here's just one of the studies done on the subject Matt:
http://www.cs.wisc.edu/wind/Publications/zfs-corruption-fast10.pdf
Are you kidding? El Reg relies on leaked documents and rumour in everything they write. This particular story just brings up the complete hypocrisy of Assanges lawyers. If you can't see the irony, then you need to have your irony bone checked by a professional.
Personally I don't see anything wrong with Assange releasing these documents. The US gov needs to be a bit more careful with who has access. It's not like Assange broke in and stole it.
Matt does have a point, but it's on the top of his head.
I'm not really sure what's wrong with T5 replacing SPARC64. SPARC is SPARC. It's not like Power or Itanic that require a recompile. It's all just SPARC. If it runs faster, then that's all just good for Oracle customers. It's like saying that IBM should not come out with Power8, cuz that means the end of Power7... what?
Who's following who into the singlethread or multithread market, who cares? IBM and Intel chose to keep pushing thread perf. Sun chose to ignore thread perf and go for many threads. Now IBM and Intel are adding more threads, while Oracle is increasing thread perf. They both end up in the same place. Who cares how they got there, other than FUD throwers like Jesper and Matt?
"So basically this price raise is only valid for new _machines_, bought after Dec 1st 2010."
Your point? So? It makes sense to protect your customers investments. I think Oracle should be applauded.
"Furthermore the licensing scheme doesn't make much sense, just look on how the 'per core' licenses for the T series processors seem to go up and down"
Seriously? You don't understand? Only a marketing droid would have trouble understanding this...
Do you really need it explained? I will explain, but I'm guessing that you really do understand... but then, what's your point?
Welcome Matt. Your snarky style has been missing from the discussion. You imply that Niagara must change it's goal of throughput to meet the further demands of single point performance. Does this mean that when Intel and IBM add cores and threads they are abandoning single point perf? Seems like a fairly naive assumption. As David points out above this has been in the roadmap for some time and the designers have been designing it for some time.
"except they do not own a single chip fab"
Really? That's your concern? So? Some would actually consider this a good thing. If the current fab has problems, use another one. IBM has the problem of supporting a relatively low volume chip manufacturing process, which must be extremely expensive for them. Manufacturing your own chips is not a feature (just ask HP).
Your other comments are just childish. We can all go over marketing terms and say we had this first, or we had it better, but it's all just nonsense. SPARC has supported domains before any of the other Unix vendors... so, who really cares?
Larry's talking about competing with NetApp, not buying them -- duh. I think it's pretty clear that when Larry said he wanted the 60% of NetAPP's business, he was talking about it going to ZFS storage. If they bought NetApp they would have to get rid of NetApp's tech or Sun's. My guess is Sun's, but I doubt that's what he'll do.
As far as buying a chip company, I bet the guy that said it will be Fujitsu's business is right. Interesting.
Now, let me see... Let's try to do this in a tidy manner:
1. When AIX 3.1 on the RS/6000 was first launced back in 1990, it wa streets ahead of any other commercial UNIX.
---This is opinion, surely, but we are all entitled to one.
2. It has a logical volume manager,
--- So did Solaris, and if I remember correctly, so did HP-UX.
3. an integrated system management utility
--- Solaris has never really had this, while HP-UX kinda has... I would argue whether
this matters, but to each his/her own.
4. dynamically loadable device drivers
--- Solaris had this when it was released (though early Solaris non-BSD was very
painful) Of course, Sun and AT&T created SVR4 (of which Solaris is based on).
5. first UNIXes that did a good job of merging SystemV with BSD flavours of commands and libraries (SUN's way of doing this was less transparent)
--- This is just opinion. I quite like the way that Sun kept the BSD equivalent commands around.
Sun, having the Father of BSD (Bill Joy) still in house made sure that the BSD advocates
could still do their thing.
6. With the SP/2 in the mid 90's IBM moved AIX into high-performance computing (Deep Blue et. al)
--- Sun purchase the business division of Cray from SGI in 1996 and released the Starfire in
1997... The SP/2 had nothing on the Starfire, especially SMP scalability. IBM was way
behind here. Sun had ignored the high-end SMP market for too long, but then went to the
leader board in one giant leap.
7. For absolutely years, AIX was the leader in the Gartner manageability surveys.
--- Marketing... The vendors pay Gartner for these "ratings"... It's amazing how often the vendor
that funds the study is the one that comes out on top... I'd bet I could go find Sun funded
studies to say the opposite.
8. Power4 systems, available in the early 2000's implemented hardware partitioning.
--- A competitive response to Sun's success with hardware domains on the Starfire system.
Sun was already into their second generation of partitioning at this time.
9. Power4 also had SMT of a kind.
--- That's like saying that hyper-threads are SMT of a kind...
10. Power5 systems, available 2004/2005 implemented I/O virtualization, sub-cpu partitioning, and dynamic hardware allocation and de-allocation.
--- Yes, IBM passed up Sun in this regard (IMHO). IBM has never really gotten the dynamic hw
allocation thing right though. Sun made up a bit for this with Zones. Personally, I think IBM's
solution has way too much overhead, but these systems are so powerful that you can
generally overlook this.
11. IBM were slow on SMP, the initial work being done by Bull with the G/J30s, but when you have systems with single CPUs running as fast as your competitors SMP boxes, what was the hurry.
--- LOL... Tell that to IBM when they were competing with the E10000 from Sun. IBM could not
compete with anything that Sun had. Even on the single proc systems... IBM had fallen
asleep and Sun was reaping all the rewards for it. IBM was slow and could not scale. That
was the facts. You can't forget that.
12. So tell me. What else were IBM lagging behind their competitors.
--- Oh yes, one other thing. Backward compatibility. Solaris, and HP-UX to a lesser extent, have
always kept compatibility between releases. IBM has always seemed to not think this was
a priority. A recompile between versions was and in many cases is a real concern with AIX.
I'm sure I have some of my "facts" mixed up above. It's mostly from memory, so it could be off by a bit, but I don't think by much.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
