All Questions

Tagged with
Filter by
Sorted by
Tagged with
0 votes
1 answer
487 views

Disk speed on Azure Virtual Machine Scale Set is slower than advertised

I have an Azure Service Fabric cluster with 5 nodes, baked by an Azure Virtual Machine Scale Set. The size of the instances is Standard_D2ds_v4 with an advertised disk speed of max 3200 IOPS. At the ...
Roman Ganz's user avatar
-1 votes
1 answer
1k views

8 x 15k SAS hdd disk with raid 0 performance vs single SSD disk for database [closed]

Which one has a better performance for database application (e.g. postgres) if we just consider read/write operations speed and not fault tolerance 8 x 15k SAS hdd disk with raid 0 single SSD disk
Ali Eb's user avatar
  • 1
-1 votes
1 answer
842 views

Hyper-V on Server 2019 - the larger the VHDX file the slower the disk speed

We have a brand new pair of servers for Hyper-V with fast RAID 10 SSDs in 2 arrays. On the host, using Crystal Disk Mark, I get speeds that I would expect like RND64K Q32T12 = 4712.49 MB/S while the ...
D_B_AINT's user avatar
8 votes
3 answers
5k views

Tmpfs vs NVME SSDs from performance POV [duplicate]

I have a script that creates large temporary files. I'm inclined to use tmpfs for this, however I did a quick search for tmpfs performance and found this which reports speed as about 2GiB/s. Modern ...
Peeyush Kushwaha's user avatar
1 vote
1 answer
654 views

How do I know if SSD is the bottleneck in Linux?

I have a database server which uses a Samsung 840 pro disk. The load is constantly higher than usual, even if there is not much activity on the website. So I suspect the disk has worn out. But how can ...
Milkyway's user avatar
  • 113
4 votes
1 answer
2k views

Slow RAID performance for Samsung EVO's

So I've been suspecting performance bottlenecks on a Samsung EVO 850 RAID1 for sometime, but honestly been too lazy to look into it. Now I was starting a new home project, involving a Vmware ESXi host ...
Sharza's user avatar
  • 51
0 votes
1 answer
10k views

Ceph hardware sizing calculator

I would like to calculate the Hardware sizing for a ceph cluster. There are so few references to the sizing, that I try to get these details here in the community. E.g. what shall i have depending on ...
cilap's user avatar
  • 277
3 votes
2 answers
1k views

Is swap compression in RAM (ZSWAP) benefitial on a Linux server with SSD drives?

I am trying to improve performance (responsivity) of a webserver during rare events of high memory utilization (90%...) where it starts to swap a lot, in order to help manual intervening faster and ...
Ján Lalinský's user avatar
5 votes
2 answers
1k views

Server Hot Swap Hard Drives

We have a HP Proliant DL380 G9 server. We have a separate data drive which consists of 4 x 1TB 7.2k 6Gb hot-swappable SAS drives RAID 5'd to provide a 3GB logical drive. We've noticed the performance ...
user avatar
1 vote
1 answer
585 views

Best Practices for Deploying Microsoft SQL Server on AWS on EC2 instance with SSD storage

Does it makes sense to install SQL Server 2017 on one single SSD drive on m5.xlarge with tempdbs, databases, program files, backups and logs for best performance? According to Amazon EBS specs the ...
mikesv's user avatar
  • 11
3 votes
1 answer
1k views

Can consumer SSD over-provisioning improve performance?

While reading up on ways to improve SSD endurance, I came across an Intel whitepaper on the subject of SSD over-provisioning (limiting the host-adressable space on a drive or not including some of the ...
kranklis's user avatar
1 vote
0 answers
261 views

Difference in utilisation reported by iostat for two identical disks in RAID1

I'm trying to debug some weird behaviour with Postgres performance and am trying to understand the iostat output. I'm on 2x SSD on RAID1. Why would one of the disks be reported at ~21% utilisation ...
Saurabh Nanda's user avatar
0 votes
0 answers
99 views

Using hdparam for more strenuous tests

So I am tasked with getting some strenuous raw hard drive performance tests. We started with FIO but this has proven to be ineffective as the results are poor in windows and linux. I tried Bonnie++ ...
LUser's user avatar
  • 217
6 votes
2 answers
1k views

Why do my SSD read latency benchmarks get markedly worse when I put an XFS filesystem on top?

I am benchmarking a small server box based on the SuperMicro E300-8D. I've installed the latest CentOS 7.5 with the latest updates, 64GB of DDR4-2100 RAM, and a Samsung 970 EVO 1TB NVMe SSD. The OS ...
anelson's user avatar
  • 305
0 votes
0 answers
532 views

SSD performance on Ubuntu + Nginx

I'm trying to improve the performance of server for nginx 100TB/month images distributions. But I have two problems: high CPU load and periodical splashes (I don't know why) Hardware is quite poor (...
demon101's user avatar
  • 103
1 vote
4 answers
513 views

Why does disk latency drop during periods of heavy writes?

Looking at the output of Munin's diskstats plugin (which is reading from /proc/diskstats) I'm noticing what seems peculiar to me. The disk is a SSD, and I would assume, magnetic or solid state, that ...
user476819's user avatar
0 votes
1 answer
580 views

MySQL perf bad on SSD-based server. Small tables. MySQL config? RAID?

I have a couple of servers in an industrial setting (air-gapped network) that do some fairly light telemetry gathering. We generate about 10GB of telemetry history in ~30 days. All telemetry goes ...
Kyle's user avatar
  • 31
6 votes
1 answer
4k views

SSD IOPS on linux, DIRECT much faster than buffered, fio

I have a 30Tb-sized hardware RAID-6 system (LSI 9280-8e) of 10 DC-S4500 Intel SSDs that is used for database purposes. The OS Debian 7.11 with 3.2 kernel. The filesystem is XFS mounted with nobarrier ...
sega_sai's user avatar
  • 185
0 votes
1 answer
122 views

GCE Persistent SSD disks allocate more IOPS than specified

I've been testing GCE and created a VM with a persistent SSD disk to test disk performance. Persistent SSD disk that I configured was 37GB, which was supposed to give me 1110 IOPS but when I run ...
user2629636's user avatar
0 votes
1 answer
771 views

Improving SQL speed - Can SSD and more RAM be effective?

I'm running a website on a VPS package on nomral Hard drive, and 3.25 GB RAM. I created a test page and realized that by simply including an SQL command such as: SELECT Name from table where Name IN ...
rockyraw's user avatar
  • 101
0 votes
0 answers
184 views

Why do I have very different read speeds at the same SSD?

In a Debian stretch, I did some quick tests for a 500GB Samsung SSD 840 read speed at all the different partitions and I cannot understand the results. My understanding so far was that the disk could ...
Kostas Papadimakis's user avatar
1 vote
0 answers
27 views

Performance decision on a 3-server HA scenario [duplicate]

I am setting up a new infrastructure for our current project, mgirating from a 2 server in 2 locations to a new 3 servers/3 datacenters... which should allow us to have a higher high availability. ...
Lorenzo Martinez's user avatar
3 votes
1 answer
10k views

HPE 1.92TB SATA 6G Mixed Use SFF SSD very slow compared to SAS HDD HP disks

In below screenshot, I benchmarked 3 machines, 4 Drives. my main production based on DL580, HPE 1.92TB SATA 6G Mixed Use SFF SSD RAID 1. MySQL insert time of 1 million rows is ~ 30 seconds in the SSD,...
Jawad Al Shaikh's user avatar
1 vote
1 answer
847 views

Intermittent I/O delays on part of RAID array causing poor MySQL performance

I am managing a bare-metal server running a MySQL database as well as some other software. We are seeing performance issues with some MySQL queries, but the issues occur intermittently, and almost ...
jnorthco's user avatar
9 votes
2 answers
8k 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 ...
SamTh3D3v's user avatar
  • 273
2 votes
1 answer
1k views

Slow SSD performance - IBM X3650 M4 (7915)

I set up a test environment for development purposes. It consists of a IBM X3650 M4 (7915) server with: 2 x Intel Xeon E2690 @ 2.90GHz 96GB 1333MHz ECC RAM 2 x HDD 146GB 15k rpm 6 x SSD 525GB (...
pietrop's user avatar
  • 179
5 votes
2 answers
20k views

Does any PCIe slot support NVMe SSD's automatically or do I need a specific M.2 slot?

I have been trying to dissect the information I find with googling, but have come to the conclusion that I cannot find with certainty the answer to the following questions: Given you know the basics ...
Abel's user avatar
  • 1,037
2 votes
2 answers
368 views

Determining if upgrading to a fast PCIe SSD drive is worthwhile on a dual Xeon Nehalem system with PCIe v2.0

I am considering upgrading my workstation (almost a server) with a bootable PCIe drive, the 2TB version of the Samsung 960 Pro. But having a system with PCI v2, SATA II I am not sure if this makes ...
Abel's user avatar
  • 1,037
1 vote
0 answers
2k views

What is the optimal strip size/full stripe size on a RAID 10 array for a database server?

I'm setting up a new SqlServer database server at work. The machine is a HP Proliant DL368 Gen9 with the SSD disks. When trying to create the array, I have to select the stripe/full stripe size. This ...
ArMaN's user avatar
  • 111
1 vote
0 answers
139 views

SSD bandwidth issue and analysis

I have Intel's P3700 SSD mounted on a Lenovo x3650 M5 server with two Intel Xeon E5-2630 v3 CPUs. The server is running Ubuntu 14.04 with 4.6.4 Kernel. I've been using fio to benchmark the SSD using ...
Gilsho's user avatar
  • 23
-1 votes
2 answers
2k views

RAID0 SSDs slow on HP ProLiant

TL;DR Is the RAID controller on the HP ProLiant DL360 G7 known to be slow or problematic? My RAID0 set is performing horribly. I have an HP ProLiant DL360 G7. It came with a 160GiB SAS HDD, and I ...
sudo's user avatar
  • 265
2 votes
2 answers
1k views

How can SSD write performance be dramatically different on two identical servers?

I am renting two identical physical servers (at Serverloft). These servers come with 2x512 GB SSD disks, and I've set them up with mdadm in a RAID1. The disks are locally installed in the server. The ...
sbrattla's user avatar
  • 1,578
3 votes
1 answer
2k views

Hardware performance penalty for Hyper-V Generation 2 VMs

I have the following configuration: Windows Server 2012 R2 OS installed on baremetal with Hyper-V. OS is installed on RAID1 850 PRO drives. Windows Server 2012 R2 Guest OS Generation 2 VM is located ...
SILENT's user avatar
  • 163
0 votes
1 answer
72 views

Which bandwith can i expect, when i´m running a SATA SSD on a SAS interface?

So, i know this question had been asked before. I consider to buy a SSD for my Server, because i kinda hate the SAS HDDs in there. I thought: If i connect a SSD, which, in fact, handles way more ...
vortexman100's user avatar
4 votes
0 answers
4k views

Tuning mdadm RAID 5 performance?

I have a system with dual Intel E5-2620 v3 @ 2.40GHz, 64 GB ram, an LSI 3008 HBA, 8x 1.6 TB Intel S3510 SSDs. I've been playing benchmarking it in different configurations using fio and I've gotten ...
Evan's user avatar
  • 487
11 votes
2 answers
18k views

does adding heaps of drives to a raid 0 increase performance?

does adding heaps of drives to a raid 0 increase performance? i know that two drives in a striped raid will usually be faster than a single drive but will i notice a difference in performance between ...
Peter Cullen's user avatar
5 votes
2 answers
10k views

Samsung 850 EVO + HP P410 limited iops?

I´m experiencing a weird write iops limitation with a HP P410 256mb cache controller and 4x consumer grade ssd´s (Samsung EVO 850) in RAID5/10. After upgrade to latest firmware (6.64) for HP P410 ...
Leo's user avatar
  • 63
2 votes
1 answer
1k views

ssd performance against hdd performance

I have two server, one with a SSD, one with a HDD. I've been running hdparm to see the differences, but the results for Timing cached reads and Timing are quite equal/HDD is faster: SSD: Timing ...
baao's user avatar
  • 123
9 votes
3 answers
32k views

Why do servers have SAS instead of SSDs? [closed]

I was wondering why do servers still come with SAS disks instead of SSD disks? I know that SAS are faster than normal hard drives but they are still much slower than SSDs. I think they are more ...
katie's user avatar
  • 101
3 votes
2 answers
1k views

ncq not supported on an SATA3.0 HDD in an usb 3.0 casing

is there a well known reason why a SATA3.0 HDD supporting NCQ does not have NCQ queue depth configured when mounted from an usb 3.0 case ? output for hdparm -iI /dev/sda shows NCQ is supported output ...
Saint Crusty's user avatar
1 vote
1 answer
955 views

6 times slower disk write in RAID10 on Ubuntu 14.04 after rebuild

OS: Ubuntu 14.04 LTS Software Raid 10: 4x120Gb Intel 530 SSDs Initial tests: 1.2 GB/s Write 5.2 GB/s Read I did some tests to check if everything will go alright with a failed disk: I set as ...
deadtired's user avatar
  • 141
1 vote
0 answers
821 views

SSD+MDRAID+LVM+KVM results in lower performance than expected

We use kvm for virtualization, the disk images are stored as logical volumes. The logical volumes are stored on a bunch of software-RAID1 (mdadm) arrays composed of Intel DCS3500 SSDs (each array is a ...
Pentium100's user avatar
1 vote
2 answers
3k views

Is it possible to use SSD for storage on Windows Azure?

There is some marketing-speak suggesting that it may be possible ..: http://gigaom.com/2012/06/05/new-windows-azure-goes-all-ssd-to-one-up-amazon-in-the-cloud/ .. but I couldn't find any way to ...
iljau's user avatar
  • 121
4 votes
3 answers
11k views

Does software RAID-0 really improve performance?

I am going to lease a server and I have two options, SATA or SSD, I decided to save the money and go for the SATA option. But I do not need all that storage they are giving me! So I thought MAYBE I ...
AL-Kateb's user avatar
  • 333
2 votes
1 answer
6k views

How to erase a SSD to restore factory performance in Linux?

Due to big performance issues with an mdraid-1 array I'd like to pull down from the array one of the devices (Samsung 840 Pro), erase it to restore factory performance and re-add it to the array. The ...
Andy B's user avatar
  • 105
2 votes
1 answer
897 views

Various problems with software raid1 array built with Samsung 840 Pro SSDs

I am bringing to ServerFault a problem that is tormenting me for 6+ months. I have a CentOS 6 (64bit) server with an md software raid-1 array with 2 x Samsung 840 Pro SSDs (512GB). Problems: Serious ...
Andy B's user avatar
  • 105
0 votes
3 answers
1k views

I/O on SSD hosting a DB periodically hangs with queue length peaks of 800+

i have a problem which i cannot identify correctly. I have a process (on WinSrv 2008 R2) which imports data into Mongo DB and periodically whole system freezes - once per 5-10-15 minutes for about 20-...
Xtra Coder's user avatar
0 votes
2 answers
1k views

Which alternative is most performant (raid, ssd, 15k rpm disks)

I'm looking at the DELL PowerEdge™ R720 from Hetzner. This server has a Dell PERC H710 8 Port SAS/SATA 6GBit/s raid controller. I have the choice between: 3 x 600GB SSD (MLC SATA 3Gb/sec) 6 x 600GB ...
rept's user avatar
  • 111
-1 votes
1 answer
2k views

Disk IO in Linux is slow (2MB/sec) while in Windows its fast (400MB/s) [closed]

I have a server with an ASUS board with a 256GB SSD In Windows, using HDTune my drive shows 400MB/sec on Linux, I'm getting about 2MB/sec using root@desktop:/# dd if=/dev/zero of=/foo bs=1024k ...
stuck's user avatar
  • 697
6 votes
2 answers
7k views

Why does storage's performance change at various queue depths?

I'm in the market for a storage upgrade for our servers. I'm looking at benchmarks of various PCIe SSD devices and in comparisons I see that IOPS change at various queue depths. How can that be and ...
Mxx's user avatar
  • 2,372