Questions tagged [benchmark]
Benchmarking is the process of gathering data about the performance of a system in different scenarios to measure the impact of changes to the system, to compare the system with other options, or to determine maximum capacity.
273
questions
9
votes
1
answer
2k
views
Understanding ZFS NVMe benchmarks with FIO
I'm trying to benchmark an all-NVMe ZFS disk array. I'm familiar with extremely fast benchmark results, with very little disk activity due to efficient ZFS caching. I ran into the reverse: massive ...
5
votes
1
answer
884
views
why read is faster when using O_DIRECT flag?
I copied a 10GB of file in my SSD which has read bandwidth of around 3.3GB/s, benchmarked using fio command. Here is the ref: https://cloud.google.com/compute/docs/disks/benchmarking-pd-performance
I ...
0
votes
0
answers
969
views
Slow read speed in ZFS mirror (slower than write speed, and very slow for small chunk sizes)
I have a server running debian on top of a ZFS 3-way mirror of Exos X18 18TB (ST18000NM001J).
I'm benchmarking it and I'm finding some surprises for the read rate under certain conditions.
But first, ...
0
votes
2
answers
1k
views
How to calculate log per second?
I have a log management system in which the Clickhouse database has been used alongside metrico/qryn, rsyslog on all the servers (Debian 11) send gathered logs from the system and applications to ...
0
votes
0
answers
110
views
Big performance gap between two similar servers in mysql-percona 5.6 ubuntu
I cannot figure out reason of performance difference between two (three?) similar percona 5.6 instances. I prepared some tpcc-like benchmarking using sysbench and among 3 similar servers, one is ...
0
votes
1
answer
2k
views
Benchmarking AWS outbound Internet bandwidth (egress) "up to 25 Gbps"
We conducted our tests on c6gn.2xlarge AWS instances located is us-east-1 region, which are advertised in AWS documentation to have a network performance of "Up to 25 Gbps" with a baseline ...
0
votes
1
answer
52
views
Measuring bandwidth from a server - determine capacity of network
I am building a system that will serve video content, and am trying to determine the maximum throughput I can achieve.
I have been using iperf3, and initially, from a separate VM (in the same data-...
0
votes
1
answer
32
views
Does Throughput calculated adds up using multiple (all active) physical adaptor
I had a network adaptor of speed 20Gbps. By stressing this adaptor using one of benchmarking tools (netperf is the tool which I used), I was able to get ~18Gbps which is well accepted.
Now, after ...
0
votes
1
answer
444
views
How to correctly benchmark sequential read speeds on 2.5" hard drive with fio on FreeBSD?
Like suggested in https://askubuntu.com/a/991311 I tested the read speed of a 2.5" spinning hard drive on a TrueNAS/FreeBSD system.
For "Sequential READ speed with big blocks":
fio --...
2
votes
1
answer
92
views
how to benchmark dbms (sql and nosql) on s390x architecture (IBM mainframes basically)
I have access to an s390x machine, a zbc12 to be precise, with 32GB of RAM, which I can use as a lab for a few months(!).
I wanted to explore this architecture capabilities, especially relating to ...
0
votes
1
answer
2k
views
Testing zpool IO performance
To test the write speed of a 2.5 inch SATA3 spinning hard drive (some old Seagate one) installed in my TrueNAS system I used dd if=/dev/zero of=/mnt/pool/test.dd bs=1024 count=1m (found this in ...
0
votes
0
answers
51
views
Benchmarking vCore Performance
A certain public cloud provider is phasing out its old small VM instances and replacing it with a new category of instances. While the old model lists "vCore 2.4 GHz" the new model features &...
1
vote
1
answer
823
views
How do I cross-reference OpenSCAP benchmarks to the CIS documentation?
I'm using the SCAP WorkBench, and have gone in to customize the CIS profile for RHEL 8. The benchmark items are clearly titled with things like "Modify the System Login Banner". The item ...
1
vote
1
answer
2k
views
How to stop apt-get using /tmp for install scripts
I have a CIS-benchmark-compliant base image.
Pulling this to differentiate an immutable image for my application, if I attempt to do
apt-get install -y docker.io
I get an error
==> amazon-ebs: Can'...
0
votes
0
answers
303
views
Benchmarking ssd MySQL write performance: sysbench oltp_write_only vs write iops, fio. Madness
I've benchmarked about 10 diffrent ssd devices with sysbench oltp_write_only.lua and I found no coroleation whatsoever between max sustained write IOPS of device (both from specification and from fio -...
0
votes
1
answer
726
views
Latency in fio vs ioping
There are two tests:
ioping -c 10 -S 4K -D .
9 requests completed in 2.49 ms, 36 KiB read, 3.61 k iops, 14.1 MiB/s
generated 10 requests in 9.00 s, 40 KiB, 1 iops, 4.44 KiB/s
min/avg/max/mdev = 259.5 ...
0
votes
1
answer
408
views
Fio brandswitch measure
In disk benchmarking tool fio there is statistics for brandswitch
Run status group 0 (all jobs):
READ: bw=20.9MiB/s (21.9MB/s), 10.4MiB/s-10.8MiB/s (10.9MB/s-11.3MB/s), io=64.0MiB (67.1MB), run=...
-2
votes
1
answer
944
views
Why Windows Server network stack is so slow compared to Red Hat Linux? [closed]
According to the following paper...
COMPARING NETWORK PERFORMANCE: RED HAT ENTERPRISE LINUX 6 VS. MICROSOFT WINDOWS SERVER 2012
Src: http://www.principledtechnologies.com/Red%20Hat/RHEL6_network_0613....
0
votes
1
answer
1k
views
fio numjobs & iodepth Values to Simulate Database PostgreSQL Workload?
I am trying to benchmark a ZFS RAID-10 array of SATA SSDs using fio with settings that are somewhat representative of database workloads like for PostgreSQL.
For example, for random reads:
fio --...
0
votes
1
answer
52
views
Will an E5-2660 be a major improvement over a Q9550 for a webserver?
I have an old production server that runs Debian on a Q9550 CPU. I want to upgrade the server but I am on a very tight budget. I can't spend more than around $600. I can get a refurbished HP ProLiant ...
0
votes
1
answer
1k
views
Where is my bottleneck: RAM, CPU or disk?
I have a 32 core, 64 thread, 128GB RAM system with internal SSD.
I'm running 64 data mining/simulation jobs in parallel that each load 1GB from disk and run for about 5-10 minutes .
I want to know ...
1
vote
2
answers
172
views
Quick Way to Evaluate Linux Server Load To Determine if Upgrade is Needed?
I've got mission critical apps running on my VDS. We're not experiencing any visible lag issues from the user perspective. However I want to be cognizant about when it's time to upgrade. The network ...
4
votes
1
answer
126
views
How can I verify my vserver provider offers SSD as promised?
For this month I have rented two vServer, each from a different provider. Both indicated that the disks of the system use SSD technology (making me expect a bandwidth of at least 300MB/s and maybe ...
3
votes
1
answer
8k
views
Litespeed vs Nginx high-traffic server performance? [closed]
Most top-level (high-traffic) websites use Nginx web servers. This community has many members who are experts about systems and servers, so I want to ask for feedback on which server can handle top-...
0
votes
1
answer
169
views
Benchmark Azure App Performance
I have an ASP.NET MVC web app running in Azure. I'm making a change to some of the resources it uses in the background but I want to be sure it's not going to have an adverse effect on the ...
0
votes
1
answer
92
views
Wordpress apache benchamrk CPU usage
I'm on a VPS, 4vCPU 4GB RAM.
During this apache benchmark test:
ab -k -n 10000 -c 3 -l URL
I get:
Server Load 3.393066 (4 CPUs)
Memory Used 35% (1,358,560 of 3,881,408)
Swap Used 23.7% (...
0
votes
1
answer
356
views
DITG (ITGSend, ITGRecv) via two network interfaces of the same machine
I have a server on which a transmission benchmarks shall be performed. This server has 3 network interfaces (172.16.10.205 for SSH, 10.0.4.105 for sending, 10.0.8.105 for receiving)
...
4
votes
1
answer
10k
views
Apache ab - testing with 1000 concurrency
I am just trying to gather some information on server capacity and ran these apache ab test.
I tested with 1, 10, 100, 1000 concurrency over 1 minute. The result below is for 1000 concurrent users
...
0
votes
1
answer
64
views
Cluster using virtual machine software with 2 workstations
I want to do a benchmark between databases (MySQL Cluster, HBase and Cassandra). I need 8 nodes. I have 2 workstations Dual-processor Z600:
1. workstation1: 16 core - 16 Go RAM
2. workstation2: 8 ...
1
vote
0
answers
15
views
Web server benchmark - explanation of results? [duplicate]
I made a small benchmark, in which I compare Apache HTTP server deployed to the Virtualbox, and Apache HTTP in Docker (official Alpine and Debian based images) deployed to the Virtualbox. The results ...
1
vote
0
answers
2k
views
Low network speed with small packets
Using iperf3 to test the network with --length 1000 --no-delay:
iperf3 --interval 1 --time 3 --no-delay --length 1000 --parallel 100 --client 10.0.0.3
Result:
[SUM] 0.00-3.00 sec 835 MBytes ...
0
votes
0
answers
201
views
How to test route lookup performance?
I need to test (IPv6) route lookup performance in the Linux kernel. I want to know how much time it takes to find a match if there are 100 routes, or 1000, or 10000 and so on. I'm using containers, so ...
2
votes
0
answers
102
views
find out NUMA locality of process RAM
I am doing an application benchmark with multiple instances of the same application. I found out that pinning their processes (with sched_setaffinity under Linux, with TaskManager under Windows) to ...
1
vote
1
answer
1k
views
Strange IOPS performance on AWS R3.large & R4.large instances
I have used 4 10GB GP2 EBS volumes with RAID0 on Windows Server 2012R2 image as explained here: http://docs.aws.amazon.com/AWSEC2/latest/WindowsGuide/raid-config.html The instance type I used was R3....
0
votes
1
answer
102
views
Memory bandwidth in OpenVZ container
What determines memory speed of OpenVZ container?
I've run sysbench --test=memory --memory-block-size=1M --memory-total-size=1G run on several OpenVZ powered VPSs from different providers and observed ...
9
votes
2
answers
9k
views
Why do sequential writes have better performance than random writes on SSDs?
An LBA (logical block addresses) is a mapping table implemented in the FTL to match between logical and physical pages/blocks in SSDs, my guess is that most SSDs (at least when they are empty) keeps ...
1
vote
1
answer
807
views
How to calculate /get the UBER value from S.M.A.R.T measurements in SSDs
UBER is a metric for the rate of occurrence of data errors, equal to the number of data errors per bits read over the entire lifetime of an SSD.
Mathematically:
The numerator is the total count of ...
4
votes
1
answer
3k
views
How to circumvent siege concurrency socket select timed out
Being trying to test load capacity of a web site. I am using an aws ubuntu xenial server t2.micro as test client and have SIEGE 3.0.8 and apache benchmark Version 2.3
while ab -c 100 -n 100000 url ...
0
votes
1
answer
2k
views
Sysbench on Redhat 7 Errors with Mysql Testing - PANIC: unprotected error in call to Lua API (cannot open oltp: No such file or directory)
When testing the mysql benchmark using sysbench on redhat 7:
sysbench --test=oltp --oltp-table-size=1000000 --mysql-db=test --mysql-user=root --mysql-password=yourrootsqlpassword prepare
I'm getting ...
2
votes
0
answers
593
views
Weird results of filesystem benchmark (bonnie++, VPS)
I wanted to check the performance of filesystems of my newly purchased VPS and compare it to the old one, the results surprised me though and I don't know how to analyse them.
I picked bonnie++ to ...
-1
votes
1
answer
186
views
Better stream performance when transparent hugepages are on
On rhel 7, when transparent hugepages are activated, I get a better stream benchmark performance than when transparent hugepages are off.
Do you have any idea of a kernel parameter (or another thing) ...
1
vote
1
answer
4k
views
Using Apache Bench with different post data per request
I'm looking to use ab to obtain some performance metrics on my site. I'm looking to send POST requests to my desired URL, however I'd like to change the values for the POST params for each request.
...
0
votes
1
answer
75
views
How do you go about comparing two servers in order to figure out how many of one, you could replace with a newer one? [duplicate]
How do you go about comparing two servers in order to figure out how many of one, you could replace with a newer one?
In particular, I have the following
Old server: ProLiant BL460c G6 (1.86 GHz, ...
0
votes
0
answers
95
views
Synology NAS X-Fer Speed Possible Issues
I recently purchased a Synology DS1815+ 8-Bay NAS for my organization and I have been running benchmarks all morning.
I purchased:
4 x 4TB 7200RPM HGST NAS (Enterprise) drives &
2 X 256GB ...
0
votes
0
answers
170
views
Strange behavior of web stress tests
After some server configuration, i decided to do a stress test using siege.
2 front servers in lxc containers, load balancing nginx - on the host machine.
(if need, I add nginx and sysctl configs ...
-3
votes
1
answer
406
views
How to configure Fio to a more accurate result of IOPs
I am using fio to test how much iops my server can offer. My conf file says to write a 20GB file and read it. THe results show iops=27291. But the test last only 10 seconds. I wonder if the test last ...
1
vote
1
answer
988
views
How to find what's causing Apache/httpd to run at such high memory usage
I'm having one site that when hit with a spider just goes off the handles. Normally everything seems fine. We have a nagios montior to report back when CPU is over 80%.
When we get the warnings, I ...
3
votes
0
answers
239
views
How to verify network quality before running important benchmarks?
I have multiple CentOS servers around the world. They all are running a PHP script that tests the latency of some services once per minute.
The problem is that sometimes these servers have local ...
1
vote
0
answers
54
views
Benchmarking websites via a browser to check loading times of a server?
I'm looking for a benchmarking tool to help find out the loading times of a website, I what to be able to run it like 250 times automatically to work out if a series of changes improves things or not.
...
5
votes
1
answer
511
views
Root-causing vastly different performance on iozone O_SYNC benchmark for two HDD manufacturers
I have two servers A and B with the following configuration:
A: 4TB HDDs, with RAID 1 (MegaRAID SAS 2008), 128MB cache, no BBU, write-through mode, 7.2k RPM, manufacturer A.
B: 1.5TB HDDs, with RAID ...