Back in the spring of this year, we transferred the infrastructure clouds mClouds.ru for fresh Xeon Gold 6254. It is too late to make a detailed review of the processor – now more than a year has passed since the release of the “stone” and the details about the processor are known to everyone. However, one feature is noteworthy, the processor has a base frequency of 3.1 GHz and 18 cores, which, when turbo-boosted, can all simultaneously operate at a frequency of 3.9 GHz, which allows us, as a cloud provider, to “upload” a consistently high frequency to virtual machines processor.
But nevertheless, we are still interested in assessing his ability under load. Let’s get started!
Brief description of the processor
As we wrote above, the processor is already familiar to everyone, but we will briefly present its specifications:
Number of Cores
CPU base clock speed
Maximum clock speed with Turbo Boost technology on all cores
Max. number of memory channels
For the test, we prepared a virtual server with 8 cores and 64 GB of memory dedicated to a virtual machine, the data is located on a fast pool based on an SSD array. Testing is performed on the Microsoft SQL Server 2014 database, the operating system is Windows Server 2016 and, of course, we cannot do without the most important thing – 1C: Enterprise 8.3 (18.104.22.1684).
We also paid attention to tests of our colleagues from Croc… If you haven’t read it, then in short: four processors were tested there – 2690, 6244 and 6254. The fastest was 6244, and the result at 6254 scored 27.62 points. We were interested in this experience, because in early tests in our cloud in the spring of 2020, we received a scatter in Gilev’s tests from 33 to 45, but it did not work less than 30, perhaps this is precisely a feature of working with another DBMS, but this prompted us to take measurements on our own infrastructure. We spent one more time and will share them.
So let’s start testing! What are the results?
As we can see, on an MSSQL server with a Xeon Gold 6254 processor with activated turbo boost, the result is 39 points. We interpret the resulting value into a Gilev score and get a result greater than the “Good” score, but not “Great” yet. We consider the result to be good in terms of evaluating this particular type of “parrots”. It is important to take into account that we did not deal with optimization at the OS and SQL server level and got the result as it is, if you wish, you can increase it a little more, but these are the subtleties of tuning, a topic for a separate blog post.
It is also worth making a reservation here that we do not call for evaluating the workload of productive databases using the Gilev test and immediately drawing conclusions about the advisability of using a particular processor, but according to our statistics, processors with a frequency of 3 GHz or more will be more efficient when working with 1C, and Gilev’s test can show different numbers, even in the conditions of one provider or in a local infrastructure. You can get high results on simpler processors, not even server ones, but this does not mean that when you “feed” the load in the form of 1C ERP for 50-100 people or Trade, you will get consistently high results. Always pilot and test if available.