Tag Info

New answers tagged

1

Dual-ported SAS disks are not intended for performance, but rather resiliency when used in multipath SAS array configurations (e.g. external JBOD or SAN situations). For single server or internal storage purposes, there's no benefit or gain achieved by using both ports. If you need better storage performance, please scale with disk count, RAID controller ...


2

RSS too is enabled in the NIC, with 8 queues. Which unfortunately did not mean that RSS was being employed, as netsh int tcp show global showed: TCP Global Parameters ---------------------------------------------- Receive-Side Scaling State : disabled After running netsh int tcp set global rss=enabled RSS started working and the load that ...


1

Having upgraded to a more recent version, I can confirm that the slow performance previously experienced is a thing of the past. The version packaged for Ubuntu 'Trusty' is over two years old (0.9.5 vs 1.2.0). Upgrading isn't too difficult except maybe for figuring out which dependencies are required (the newer version seems to have quite a few more than ...


3

Yes, enterprise SAS disks spin at a fixed rotation speed. They will not go into power or energy-saving modes like certain consumer drives.


0

The 20 second delay could also be caused by IIS having to restart it's w3wp.exe which will go to sleep when it's unused.


1

Some great tips here. For me the thing that made the biggest difference (from 8 seconds to 0 on IIS 7) was enabling caching and changing $wgDBserver = "127.0.0.1"; (from localhost). For some reason localhost was being resolved via the network. Here is the whole post by Jaded: One of the biggest impacts, reducing the load from 15 seconds down to around 6 ...


4

On a Windows host, qemu isn't actually a hypervisor, but is doing full machine emulation with dynamic translation, which is horrendously slow, and there's little that can be done to speed it up. It's maybe useful as a demonstration or for debugging purposes, but for anything serious you will want to use an actual Windows hypervisor such as Hyper-V, or some ...


1

I had similar symptoms, not on a VM but a Win7 Laptop. Taskmgr.exe and other processes (like Winmerge.exe) were taking 13% of cpu (one out of 8). The issue was temporarily alleviated by restart, so I was doing that a lot. After long time investigating it, I figured that that Webroot Secure Anywere (software for virus and other types of protection) ...


0

Just a wild guess. Your problem may be big dirty page flushing. Try to set up /etc/sysctl.conf like: # vm.dirty_background_ratio contains 10, which is a percentage of total system memory, the # number of pages at which the pdflush background writeback daemon will start writing out # dirty data. However, for fast RAID based disk system this may cause large ...


5

That particular RAID controller claims 8-lane PCI Express 2.0 compliancy meaning that effectively you'll already be limited to 8 * 4 Gbs = 32 Gbs (or 4000 MB/s) regardless of what's connected to the RAID card. Each SAS SFF8088 connector will cary 4 SAS lanes over a single cable, when each link is at the maximum 6 Gbs port speed theoretically you indeed get ...


3

Based on my experience with similar controllers like Smart Array P420i from ProLiant ML350p Gen8 or ServeRAID M5015 SAS/SATA Controller from IBM x3560 M3 I'd say: 6Gbps per mini-SAS (i.e. 12Gbps) - best possible scenario. I have not tested two SFF8088 ports, only one, and I've hit 6 gbps limit.


-2

I work for a Church/School. We had some Linksys routers in the classrooms because all of the students use laptops and bring a smart phone. There was an average of 30-40 students on each router. After 8 months, the router would suddenly only run 5 mb/s wired and wireless. they were mounted on 8 foot ceilings with a/c running in the rooms. The only one ...


0

I know this is old; but I'm mainly replying for users who come here with similar issues. There are a few ways to check interrupts: cat /proc/interrupts records interrupts per CPU and IO device. grep cpu /proc/stat column 6 and 7 contain irq and softirq stats. top and htop; observe si and if you change htop to advanced display mode for CPUs you can see si ...


4

Your storage admin isn't giving you the whole story. IOPS are what you will be limited by when you're doing normal IO (random seeky reads and small block writes), however you're correct that backups should be a big sequential read and can't be measured helpfully in IO/s. When you do a backup, are you reading contiguously, or are you doing lots of small ...


3

There is no problem. This is normal and expected behaviour. The reason for the packet loss is that UDP doesn't have any congestion control. In tcp when congestion control algorithms kick in, it will tell the transmit end to slow down the sending in order to maximise throughput and minimise loss. So this is entirely normal behaviour for UDP actually. UDP ...


0

You completely missed out on the most obvious culprit - the adapters, and then the drivers (you state that using a different version drivers yields different results). Try turning all offload capabilities off.


1

Inntran's is right here: you can't use VM acceleration without loading a kernel module/driver, which need administrative privileges. However, this seems a bad method of software distribution to mee: you don't need to deploy a Ubuntu images on every client, you simply need to install a single, shared instance and use it via SSH (for text-only programs) or ...


2

RAID5 is unsuitable full stop A good explanation (from 5 years ago) is here: http://storagemojo.com/2010/02/27/does-raid-6-stops-working-in-2019/


1

I use a bunch of EC2 instances; I've also seen wide variance in how long the same job would take to complete - processing the same video again and again. Sometimes, two runs are very fast, variance less than 1%; other runs the variance is 5 to 10x! I've noticed that these issues occur more on our smaller instances. We have a few that are the M*.large or ...


2

depending on your instance type (burstable cpu instances), you may be dealing with CPU credit. ==> http://stackoverflow.com/questions/28984106/whats-is-cpu-credit-balance-in-ec2 if you find vast fluctuations on a non-burstable instance, you may want to stop and start (not reboot) instance. this will make the instance come up on a different hypervisor ...


1

I suspect the receiving side is simply not capable of handling the packet rate and here's why: using tcpdump on the client reduces the packets dropped: tcpdump is slowing down the client and therefore the server is seeing a much lower packer rate which it can still partially handle. You should be able to confirm this hypothesis by checking the RX/TX packet ...


1

Another way is ip_conntarck module, Are you sure your linux-box can accept new connection? test via : root@debian:/home/mohsen# sysctl net.ipv4.netfilter.ip_conntrack_max net.ipv4.netfilter.ip_conntrack_max = 65536 root@debian:/home/mohsen# sysctl net.ipv4.netfilter.ip_conntrack_count net.ipv4.netfilter.ip_conntrack_count = 29 You have to test ...


9

When tcpdump is running, it will be fairly prompt at reading in the incoming frames. My hypothesis is that the NIC's packet ring buffer settings may be a bit on the small size; when tcpdump is running it is getting emptied in a more timely manner. If you're a Red Hat subscriber, then this support article is very useful Overview of Packet Reception. It has ...


2

What power governor are you using? I've seen similar behaviors with "ondemand" or "conservative" governor. Try to use the "performance" governor and to disable any powersaving features in the server BIOS. Does it change something?


0

What are the specifications of your Microsoft SQL Server, i.e. RAM, Disk, and CPU? How many Microsoft CRM Users do you have (both licensed and typical concurrent)? While not typically a factor in slow workflow performance, if the Microsoft CRM Server is underpowered, it could affect how fast the workflows are processed too. What are the hardware ...


0

No, there are no better options. I mean you can get more beast out of your system if you use RAID 0 (stripe), but that is not a clever idea under a database server, because in RAID 0 if only one disk fails: you lost everything. So, create a RAID 1 device, shut down database, copy over the files, check the permissions, and mount the new device under ...


2

By default, the value of parameter smtp_host_lookup is dns. Because of that, postfix will always try to resolve the MX record recipient domain to determine the next-hop destination. Postfix rely on libc resolver to do the lookup so the expected behaviour would depends on the library. For example, postfix will alwasy do lookup via NS server defined in ...


0

Looking at your details, we can conclude these general recommendations : Your READ : WRITE ratio (61 : 39 ) gives the clue that your engine types should be changed to Innodb. Before that check the output of SHOW ENGINE INNODB STATUS\G; SHOW FULL PROCESSLIST; during peak hours. Most probably you will see queries in LOCK state. Table locks is the one big ...


3

Your host does not seem to be caching dns lookups. It should be. No Your operating system (not postfix) should cache every DNS record lookup it needs. NSCD is one common tool to cache "hosts" queries.


1

The answer is that it really depends on your specific environment. Without many more details, no one can answer this for you. However we can provide some guidance that may help you in making this decision. Each infrastructure or AP wireless device you have on a channel (in the same area) will increase the amount of "air time" on that channel used by ...


1

Connect to your existing network, without question. Not only will this reduce RF noise, but it will also greatly simplify the instructions you'll need to provide to the users of these devices.


1

I have never experienced issue such as long load times like that, however one thing I have noticed that tend to speed up my Wordpress load times is to use WP Super Cache. It caches the pages as plain HTML pages for most users, that way your server doesn't need to recompile the page every time. As others have said, it could be poor server performance on ...


1

I'm afraid your experience is clearly insufficient to deal with such issues. You probably should hire someone to either supervise the server for you, or at least get it straight now (and get rid of cPanel). He might even tell you it is not a server issue, but rather WordPress thing. You should start with looking at the load itself (CPU, memory, I/O, etc., ...


0

We can tell only one thing. Change of ISP if user complain as we don't know the load on your website. (& We dont see an important stat, the UPLOAD) On my own ISP, far less than your speed I host a small site, really enought for it, it all depend on your usage.


2

To me it looks like that the plugin does simply so heavy processing for every request that the requests take such a long time. You should ask the plugin provider if they can explain the slowdown in any way.


1

Bit late to the party here, but for what it's worth "attached" SSDs in Azure have their IOPs throttled based on the machine size. It wasn't mentioned in the pricing anywhere but I raised this issue as a ticket with technical support when they referred me to the blog post below. See this link: ...


6

You might know what a core is from shopping for desktops and laptops. It's the CPU, the processor, the bit that does the work. Here's a wikipedia link that might help. I plugged your site into this tool and most of the complaints were CSS and javascript that had to fully download before your site could fully render. Of course, I imagine we're viewing the ...



Top 50 recent answers are included