We fully understand the different versions available. The version used was clearly documented in the report. Had we not mentioned the verion used, or implied that this was NOT the free version, then we would have "screwed up".
Just because we didn't test the version that you would like us to have tested does not mean that the test is "screwed up". Please feel free to run your own tests on ESX Server if that is what you are interested in.
We ran the tests as designed because (1) We have 2 VMWare Server (free version) running in our office and were curious about the performance impact under load (2) We believe that there are a lot of smaller organizations doing the same thing
In the future, I recommend that you ignore evaluations that are not relevant to the needs of your organization.
No, actually, we didn't f**k up our setup. To quote from the article:
"The virtualized server has the same memory available to it (2G) as the native server (which implies that the physical machine running VMware has more memory)."
Apparently this sentence is confusing to some, maybe because it was not obvious that we had two completely separate machines in the test setup. I'll try to clarify: When the physical server was tested, it had 2G. When the virtualized server was tested, it had 2G. The physical machine running the virtualized server has _more_ than 2G. We should have stated how much - it had 4G, all of which was available to the host OS. I hope that clears it up.
While I applaud Google for their efforts in the past, this change seems to have thrown out the baby with the bathwater.
As an example, consider the search phrase "load testing software" (one which I have particular interest in). Under the new algo, the 5th entry is a download page of a website that sells software, but not load testing software. The word "testing" does not appear on that page at all...and the word "load" appears only as part of the word "download". Entries that are much more relevant (such as company home pages that sell "load testing software") appear much farther down the page. The phrase "load testing software" doesn't appear anywhere in that entire website, but other sites, which do have that exact phrase, do not appear on the first two pages.
I fail to see how that helps me find what I'm looking for. I appreciate Google's efforts at reducing illegitimate searches, but they seem to be off the mark this time.
Slashdot Mirror
We fully understand the different versions available. The version used was clearly documented in the report. Had we not mentioned the verion used, or implied that this was NOT the free version, then we would have "screwed up".
Just because we didn't test the version that you would like us to have tested does not mean that the test is "screwed up". Please feel free to run your own tests on ESX Server if that is what you are interested in.
We ran the tests as designed because
(1) We have 2 VMWare Server (free version) running in our office and were curious about the performance impact under load
(2) We believe that there are a lot of smaller organizations doing the same thing
In the future, I recommend that you ignore evaluations that are not relevant to the needs of your organization.
No, actually, we didn't f**k up our setup. To quote from the article: "The virtualized server has the same memory available to it (2G) as the native server (which implies that the physical machine running VMware has more memory)." Apparently this sentence is confusing to some, maybe because it was not obvious that we had two completely separate machines in the test setup. I'll try to clarify: When the physical server was tested, it had 2G. When the virtualized server was tested, it had 2G. The physical machine running the virtualized server has _more_ than 2G. We should have stated how much - it had 4G, all of which was available to the host OS. I hope that clears it up.
While I applaud Google for their efforts in the past, this change seems to have thrown out the baby with the bathwater.
As an example, consider the search phrase "load testing software" (one which I have particular interest in). Under the new algo, the 5th entry is a download page of a website that sells software, but not load testing software. The word "testing" does not appear on that page at all...and the word "load" appears only as part of the word "download". Entries that are much more relevant (such as company home pages that sell "load testing software") appear much farther down the page. The phrase "load testing software" doesn't appear anywhere in that entire website, but other sites, which do have that exact phrase, do not appear on the first two pages.
I fail to see how that helps me find what I'm looking for. I appreciate Google's efforts at reducing illegitimate searches, but they seem to be off the mark this time.