tag:blogger.com,1999:blog-2825935617443597949.post2208157758772830419..comments2024-03-22T02:10:43.183-07:00Comments on The Cloud Architect's Blog: Are Commercial J2EE Application Servers worth their cost?Derek C. Ashmorehttp://www.blogger.com/profile/02379227156744675210noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-2825935617443597949.post-86132349214176974382014-01-26T19:35:09.256-08:002014-01-26T19:35:09.256-08:00Commenting on your blog is not wasting time, it...Commenting on your blog is not wasting time, it's a privileged to every reader. http://homeextensioncentre.co.ukAnonymoushttps://www.blogger.com/profile/02168809493352908079noreply@blogger.comtag:blogger.com,1999:blog-2825935617443597949.post-70108352589018074882013-12-13T19:53:05.480-08:002013-12-13T19:53:05.480-08:00Thank's Derek ..I like your way of sharing it ...Thank's Derek ..I like your way of sharing it helps a lot !! <br />http://www.tateberry.com<br />Anonymoushttps://www.blogger.com/profile/02168809493352908079noreply@blogger.comtag:blogger.com,1999:blog-2825935617443597949.post-37310504671577091232012-10-11T04:52:25.335-07:002012-10-11T04:52:25.335-07:00Thanks for the blog reference. That reference als...Thanks for the blog reference. That reference also confirms that the survey consists of purely New Relic customers. I agree that share for Weblogic and Websphere are likely under reported and JRun market share over reported. I do think it unlikely that Weblogic/Websphere market share is so under reported that they are in fact have higher market share than open source alternatives. I do not believe that architects should ignore market share reports; but we do have to weight those reports accordingly given report limitations.<br /><br />Looking for other sources, I also found a 2011 survey (http://zeroturnaround.com/java-ee-productivity-report-2011/) that had weblogic/websphere at a combined market share of 17%. What was interesting about that report, which I'm sure has its own problems, is that taken together with their 2009 report (http://zeroturnaround.com/labs/java-ee-container-redeploy-restart-turnaround-report/) shows a substantial decrease in commercial app server usage and increase in open source app server usage.<br /><br />Anyway, thanks for a most interesting discussion.Derek C. Ashmorehttps://www.blogger.com/profile/02379227156744675210noreply@blogger.comtag:blogger.com,1999:blog-2825935617443597949.post-19596622937502797572012-10-10T18:24:19.249-07:002012-10-10T18:24:19.249-07:00Sorry, I didn't mean to imply that New Relic w...Sorry, I didn't mean to imply that New Relic was intentionally being misleading about their data or being dishonest in any way. All I'm saying is that we as architects shouldn't draw any serious conclusions about this data in terms of overall marketshare of commercial application servers. Are we really to believe that JRun has a greater market share than both Weblogic and Websphere?<br /><br />I found a post discussing the study in more detail.<br />http://blog.newrelic.com/2012/06/13/new-relic-redmonk-state-of-the-stack/<br /><br />"it’s necessary to state up front that RedMonk is not asserting that New Relic’s data is statistically representative of the market as a whole, or even of the specific communities surveyed."Ken Liuhttp://www.kenliu.netnoreply@blogger.comtag:blogger.com,1999:blog-2825935617443597949.post-21544362254851067572012-10-10T00:16:42.414-07:002012-10-10T00:16:42.414-07:00Thanks for reading my blog entry and taking time t...Thanks for reading my blog entry and taking time to comment.<br /><br />Your point is well taken that the source of the data needs to be looked at. I referenced just the graphic, but they claim (yes, this could be a misrepresentation) the info was produced from an independent survey (ref below - see second paragraph in the linked page). They don't state how the survey participants relate to their customer base. Perhaps you know more about the source of this data than what was published.<br /><br />Not having any inside knowledge of the firm, I look at motives. A firm in the monitoring business would have use for independent market research of this type. According to their documentation, they monitor at a low enough level to publish a list of specific application servers they support. Consequently, this type of market research would be very valuable for them since it tells them where to invest their R&D money. Using intentionally skewed survey results would simply lead to mis-allocation of R&D. Also, I have trouble imagining their motive for intentionally publishing a mis-leading survey.<br /><br />Having said that, better disclosure of the survey process would have been useful. <br /><br />Ref: http://siliconangle.com/blog/2012/01/11/open-source-java-is-stronger-than-ever-infographic/Derek C. Ashmorehttps://www.blogger.com/profile/02379227156744675210noreply@blogger.comtag:blogger.com,1999:blog-2825935617443597949.post-16074279128161264232012-10-09T17:42:48.225-07:002012-10-09T17:42:48.225-07:00Hi Derek -
I fully agree in your overall premise ...Hi Derek -<br /><br />I fully agree in your overall premise that commercial app servers are not a good choice.<br /><br />However, I believe the market study you cited represents a VERY skewed data set. I don't believe their customer data reflects the whole application server market. New Relic is a cloud monitoring software vendor which fills a gap open for Tomcat users who do not have the sophisticated management tools available to Weblogic/Websphere users. New Relic has made a name for itself among the Ruby on Rails crowd, which is a much different set of users than the "Enterprise" customers who are on Weblogic.<br /><br />KenKen Liuhttp://www.kenliu.netnoreply@blogger.com