The Giga Information Group released a report on Dec. 12 containing the results of their web-based survey on app. server usage. The Survey gave BEA 56% marketshare, 33% for IBM, 3% for IPlanet, and 7% for all others combined.
The report was based on a web-based survey of over 1000 people, and the report cautions that its results "should be read as more of a popularity contest than a true measure of use or sales penetration".
- Giga on App. Server Marketshare: BEA: 56%, IBM: 33%, IPlanet: 3% by Rasmus Ravn on January 11 2001 12:02 EST
- Giga on App. Server Marketshare: BEA: 56%, IBM: 33%, IPlanet: 3% by Veena Pam on January 11 2001 23:35 EST
- Giga on App. Server Marketshare: BEA: 56%, IBM: 33%, IPlanet: 3% by Floyd Marinescu on January 12 2001 13:53 EST
- Giga on App. Server Marketshare: BEA: 56%, IBM: 33%, IPlanet: 3% by Stu Charlton on January 14 2001 07:20 EST
- Giga on App. Server Marketshare: BEA: 56%, IBM: 33%, IPlanet: 3% by Anye S. on January 16 2001 13:33 EST
- Giga on App. Server Marketshare: BEA: 56%, IBM: 33%, IPlanet: 3% by Jaya Prakash E.Subramanian on January 16 2001 14:45 EST
- Giga on App. Server Marketshare: BEA: 56%, IBM: 33%, IPlanet: 3% by Ron Poon on January 16 2001 21:49 EST
Here is a response from Dave Wolf taken from a Sybase Newsgroup (Repeated)
Funny, the report from GIGA I have from threes weeks ago is a correction that they were incorrect in believing that iPlanet held a #3 spot, and believe that spot is held by market leading vendors like Sybase.
Quoted directly from GIGA:
""In August 1999, Giga published Comparison of Three EJB Application Server Solutions: IBM, BEA Systems and iPlanet (Sun/Netscape) We chose to include the iPlanet application server in that comparison because, at the time, we expected it to be a clear third in market share compared to IBM and BEA. So far, that has almost, but not quite, happened, due to iPlanet's entry into the market a year or more later than BEA and IBM, and we are projecting that iPlanet will take a 9 percent share of the 2000 market, considerably smaller than IBM and BEA. Still, among the vendors capable of moving closer to the IBM or BEA level of market share in 2001, iPlanet is one of the leaders, along with Sybase ....." -- The GIGA Group
I had a different report, indicating 24% market share for both BeA and IBM with their app server products at the end of Q4 - 2000. IBM has a grip on the Asia Market, exceeding its own expectations.
Others have not crossed double figures for marketshares.
Can you reveal the source of the report?
IBM' s marketshare went up, per IBM's own expected rise, and the source is a journal called "Computers Today" based out of India. I do not have the actual copy, but I've read the article and they have a neat pie-chart. It's simply self-evident that the market for App Servers in not significant compared to the US/Europe, due to purchasing power parities. Hence the market report is world-wide and shows [surprise, surprise] 24% for each BeA and IBM. The next is a distant 9% for iplanet.
BeA scores in tech support and IBM's scores in web resources/redbooks based support. IBM's products are completely java-based and their IBM Visual Age is the best, and I was recommended this IDE during my training by my trainer from Oracle! Besides some problems with Wesbphere include not finding the solution ready, while it actually exists and developers never, ever read documentation and play it on the vendor, be it whoever.
e.g Charles Schwab?
"This could be especially true considering that the BEA contract doesn't displace archrival IBM (IBM:NYSE - news) at Schwab, counter to BEA's take on the deal."
If you have some other market share figures, please do let me know. :-)
I cant say i'm particulary excited about tomcat, but JBoss certainly seems to be commercially viable. I wonder if the proliferation of these products "free" (depending on licence of cours) into the mainstream will errode market share, something like how http servers once costed thousands of dollars but are now ridiculously easy to write and given away for free.
I think that open source products like JBoss compete primarily with the "other" category as opposed to BEA & IBM. This is because these super big corporations will pay for commercial software licences. My prediction for next year would look like 55% BEA, 35% IBM, 5% JBoss, 5% other.
Would somebody provide for the exact figures for App Server Market share? This year, per several reports, BeA WL has lost share while IBM WAS has moved up. Others including iplanet, Dynamo, Sybase etc have remained around the same.
The report from GIGA is conflicting with the other analyst reports that I read.... IDC reports cost $3500 and that's a lot of dough [DUH!]
Sorry guys, the original post said IBM had 33% and then 3%. In fact IPlanet had 3%. My typo. :)
Strange that Websphere seem to have such a high market share when it nobody really seem to think it's a good EJB-server. But I guess IBM is a strong brand-name. *sigh*
I think IBM pricing has reflected in its % share of the market, thought IMO Websphere is not as good an appserver as to capture that big a share. Also, I would really doubt any claims that iPlanet holds 3% of the market share
When we started our J2EE development 8 months ago, we debated whether to go with WebLogic or Websphere. Finally, VAJ and WebLogic's development price tilted the balance in favour of Websphere. We have had very good luck with WAS. It's stable and fast (although not been stress tested yet). Most importantly, VAJ makes RAD a reality.
Apart from the fact that it's behind the curve on the EJB specs, which may be a big deal for some, why do you feel that it's not a good EJB server?
I'm not too surprised by this survey, in terms of *relative* shares. The exact numbers I'm a little skeptical of.
i-Planet used to have a lot of market share in the KIVA days, that could be why it has a relatively high market share.
Another thought is that they're ignoring rollouts of Microsoft Transaction Server / COM+ and non-J2EE app servers such as Apple WebObjects, which would give the "other" category a bigger share. WebObjects was in 1998 the #1 app server by number of rollouts. These days, there seems to be an unwritten rule by analysts and trade rags to completely ignore this offering, even though it was voted JDJ's #1 app server for 2 years in a row, and Network Computing's best product for 1999.
What's amazing is BEA's marketing genius. These guys have just executed *flawlessly*. People don't seem to talk about J2EE anymore, they talk about WebLogic.
The question is, does the technology reflect the use? I really do like WebLogic; but I don't think its technology is significantly superior than offerings from, say, GemStone or Persistence. Definitely better than stuff from IBM -- which is still surprising to me that they've dropped the ball on WebSphere so badly, yet were instrumental in defining J2EE.
One of the old thoughts was that WebLogic Enterprise (formerly M3) would be the "real contender" for beefy systems, since it integrates Tuxedo. WebLogic would be for "front line" systems, and WL Enterprise was for the backoffice. BEA spent *years* perfecting this product before they bought WebLogic. But I don't see this server being marketed very well -- big sites such as Amazon.com use it, but most people don't even know about it, or why you would use it over plain vanilla WebLogic.
WebObjects is a couple of years ahead of any j2ee application server today. it's been around for 10 years!
the reason they're not including it in their reports, because the other app server are simply not there. it's definitelly #1 in terms of price and work efficiency (once you become familiar with it :)
just my 2 cents
What? No ATG Dynamo?
I see lots of folks working with Dynamo, those that have used BEA, IBM, and ATG far prefer the latter.
I personally haven't worked with it yet, but have worked with both WebLogic and WebSphere (and prefer WebLogic of those two).
I think your survey was only in West coast.when you compare application server is east coast IBM have market compare to weblogic and iPlanet.
In the presentations offered by BEA(HK) and Sybase(HK) three months ago, they showed simliar market share pie-charts - WebLogic 32%, WebSphere 16%, Sybase 15%. And as far as I know, that 32% including both WebLogic Server(formerly Tengah(bought by BEA), 100% Java-based, EJB) and WebLogic Enterpirse(TUXEDO, C-based, CORBA). I am not sure whether the figures are correct as there is full of propaganda in the market. I am not sure how much WebLogic Server accouts for that 32%, if the figure is correct. Do you have any idea?
Why are you all getting so hung up on market share ? These figures are meaningless, as they don't seem to mention which market. If you focused on large-scale deployments to high load sites, I'd bet my bottom dollar that the iPlanet share would go up and that Dynamo would make it into the top 3 (maybe 4).
I got into this loop due to an article that I read on a long flight over the Pacific.
[you gotta search here]
I hear BeA WL lost 50% share and Websphere moved up a little bit and iplanet or ATG came up third [not sure which]. So which app server would I focus on - BeA 6.1 or WAS 3.5 [their site mentions 4.0 is due etc].
Guess these are both good and would stay that way for sometime - unless the third player grabs marketshare...........
I feel that market share is not a reflection of which product is best. If anything it just reflects which products are most widely (and cheaply) available. Have a look at :
I serious doubt the validity of the forrester report, when I see they put JRun in the same league as iPlanet. It's just a servlet/jsp engine. I think people here are talking about the Java based app server.
According to research from Giga, IDC and Gatner, BEA has more than double the market share of the application server market than any competitor.
The Forrester research was licenced from Doculabs (www.doculabs.com) who did not include BEA WebLogic in the survey becuase, at the time, BEA were just about to ship WebLogic 6.0 and did not want to have the previous version included in the review.
I can't but agree with you. What's the big deal about Market share which is basically how many developers have got the freebie downloads. I wonder how that affects the price of fish in the market or how many live site are running using it. Perhaps more importantly, is the Online Business getting any return of investment from the site.
The sooner people get away from all the religious arguements about which App Server is what and concentrate on what feature set each of this vendors can deliver to a business going online. The more Job Statisfaction we will all have. I love this industry, but it is the most childish industry in the world