Jul
13
2018
--

On MySQL and Intel Optane performance

MySQL 8 versus Percona Server with heavy IO application on Intel Optane

Recently, Dimitri published the results of measuring MySQL 8.0 on Intel Optane storage device. In this blog post, I wanted to look at this in more detail and explore the performance of MySQL 8, MySQL 5.7 and Percona Server for MySQL using a similar set up. The Intel Optane is a very capable device, so I was puzzled that Dimitri chose MySQL options that are either not safe or not recommended for production workloads.

Since we have an Intel Optane in our labs, I wanted to run a similar benchmark, but using settings that we would recommend our customers to use, namely:

  • use innodb_checksum
  • use innodb_doublewrite
  • use binary logs with sync_binlog=1
  • enable (by default) Performance Schema

I still used

charset=latin1

  (even though the default is utf8mb4 in MySQL 8) and I set a total size of InnoDB log files to 30GB (as in Dimitri’s benchmark). This setting allocates big InnoDB log files to ensure there is no pressure from adaptive flushing. Though I have concerns about how it works in MySQL 8, this is a topic for another research.

So let’s see how MySQL 8.0 performed with these settings, and compare it with MySQL 5.7 and Percona Server for MySQL 5.7.

I used an Intel Optane SSD 905P 960GB device on the server with 2 socket Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz CPUs.

To highlight the performance difference I wanted to show, I used a single case: sysbench 8 tables 50M rows each (which is about ~120GB of data) and buffer pool 32GB. I ran sysbench oltp_read_write in 128 threads.

First, let’s review the results for MySQL 8 vs MySQL 5.7

After achieving a steady state – we can see that MySQL 8 does not have ANY performance improvements over MySQL 5.7.

Let’s compare this with Percona Server for MySQL 5.7

MySQL 8 versus Percona Server with heavy IO application on Intel Optane

Percona Server for MySQL 5.7 shows about 60% performance improvement over both MySQL 5.7 and MySQL 8.

How did we achieve this? All our improvements are described here: https://www.percona.com/doc/percona-server/LATEST/performance/xtradb_performance_improvements_for_io-bound_highly-concurrent_workloads.html. In short:

  1. Parallel doublewrite.  In both MySQL 5.7 and MySQL 8 writes are serialized by writing to doublewrite.
  2. Multi-threaded LRU flusher. We reported and proposed a solution here https://bugs.mysql.com/bug.php?id=70500. However, Oracle have not incorporated the solution upstream.
  3. Single page eviction. This is another problematic area in MySQL’s flushing algorithm. The bug https://bugs.mysql.com/bug.php?id=81376 was reported over 2 years ago, but unfortunately it’s still overlooked.

Summarizing performance findings:

  • For Percona Server for MySQL during this workload, I observed 1.4 GB/sec  reads and 815 MB/sec  writes
  • For MySQL 5.7 and MySQL 8 the numbers are 824 MB/sec reads and  530 MB/sec writes.

My opinion is that Oracle focused on addressing the wrong performance problems in MySQL 8 and did not address the real issues. In this benchmark, using real production settings, MySQL 8 does not show any significant performance benefits over MySQL 5.7 for workloads characterized by heavy IO writes.

With this, I should admit that Intel Optane is a very performant storage. By comparison, on Intel 3600 SSD under the same workload, for Percona Server I am able to achieve only 2000 tps, which is 2.5x times slower than with Intel Optane.

Drawing some conclusions

So there are a few outcomes I can highlight:

  • Intel Optane is a very capable drive, it is easily the fastest of those we’ve tested so far
  • MySQL 8 is not able to utilize all the power of Intel Optane, unless you use unsafe settings (which to me is the equivalent of driving 200 MPH on a highway without working brakes)
  • Oracle has focused on addressing the wrong IO bottlenecks and has overlooked the real ones
  • To get all the benefits of Intel Optane performance, use a proper server—Percona Server for MySQL—which is able to utilize more IOPS from the device.

The post On MySQL and Intel Optane performance appeared first on Percona Database Performance Blog.

Jul
11
2018
--

AMD EPYC Performance Testing… or Don’t get on the wrong side of SystemD

Ubuntu 16 AMD EPYC

Ever since AMD released their EPYC CPU for servers I wanted to test it, but I did not have the opportunity until recently, when Packet.net started offering bare metal servers for a reasonable price. So I started a couple of instances to test Percona Server for MySQL under this CPU. In this benchmark, I discovered some interesting discrepancies in performance between  AMD and Intel CPUs when running under systemd .

The set up

To test CPU performance, I used a read-only in-memory sysbench OLTP benchmark, as it burns CPU cycles and no IO is performed by Percona Server.

For this benchmark I used Packet.net c2.medium.x86 instances powered by AMD EPYC 7401P processors. The OS is exposed to 48 CPU threads.

For the OS I tried

  • Ubuntu 16.04 with default kernel 4.4 and upgraded to 4.15
  • Ubuntu 18.04 with kernel 4.15
  • Percona Server started from SystemD and without SystemD (for reasons which will become apparent later)

To have some points for comparison, I also ran a similar workload on my 2 socket Intel CPU server, with CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz. I recognize this is not most recent Intel CPU, but this was the best I had at the time, and it also gave 48 CPU Threads.

Ubuntu 16

First, let’s review the results for Ubuntu 16

Or in tabular format:

Threads Ubuntu 16, kernel 4.4; systemd Ubuntu 16, kernel 4.4;

NO systemd

Ubuntu 16, kernel 4.15
1 943.44 948.7 899.82
2 1858.58 1792.36 1774.45
4 3533.2 3424.05 3555.94
8 6762.35 6731.57 7010.51
12 10012.18 9950.3 10062.82
16 13063.39 13043.55 12893.17
20 15347.68 15347.56 14756.27
24 16886.24 16864.81 16176.76
30 18150.2 18160.07 17860.5
36 18923.06 18811.64 19528.27
42 19374.86 19463.08 21537.79
48 20110.81 19983.05 23388.18
56 20548.51 20362.31 23768.49
64 20860.51 20729.52 23797.14
72 21123.71 21001.06 23645.95
80 21370 21191.24 23546.03
90 21622.54 21441.73 23486.29
100 21806.67 21670.38 23392.72
128 22161.42 22031.53 23315.33
192 22388.51 22207.26 22906.42
256 22091.17 21943.37 22305.06
512 19524.41 19381.69 19181.71

 

There are few conclusions we can see from this data

  1. AMD EPYC CPU scales quite well to the number of CPU Threads
  2. The recent kernel helps to boost the throughput.

Ubuntu 18.04

Now, let’s review the results for Ubuntu 18.04

Threads Ubuntu 18, systemd
Ubuntu 18, NO systemd
1 833.14 843.68
2 1684.21 1693.93
4 3346.42 3359.82
8 6592.88 6597.48
12 9477.92 9487.93
16 12117.12 12149.17
20 13934.27 13933
24 15265.1 15152.74
30 16846.02 16061.16
36 18488.88 16726.14
42 20493.57 17360.56
48 22217.47 17906.4
56 22564.4 17931.83
64 22590.29 17902.95
72 22472.75 17857.73
80 22421.99 17766.76
90 22300.09 17773.57
100 22275.24 17646.7
128 22131.86 17411.55
192 21750.8 17134.63
256 21177.25 16826.53
512 18296.61 17418.72

 

This is where the result surprised me: on Ubuntu 18.04 with SystemD running Percona Server for MySQL as a service the throughput was up to 24% better than if Percona Server for MySQL is started from a bash shell. I do not know exactly what causes this dramatic difference—systemd uses different slices for services and user commands, and somehow it affects the performance.

Baseline benchmark

To establish a baseline, I ran the same benchmark on my Intel box, running Ubuntu 16, and I tried two kernels: 4.13 and 4.15

Threads Ubuntu 16, kernel 4.13, systemd Ubuntu 16, kernel 4.15, systemd
Ubuntu 16, kernel 4.15, NO systemd
1 820.07 798.42 864.21
2 1563.31 1609.96 1681.91
4 2929.63 3186.01 3338.47
8 6075.73 6279.49 6624.49
12 8743.38 9256.18 9622.6
16 10580.14 11351.31 11984.64
20 12790.96 12599.78 14147.1
24 14213.68 14659.49 15716.61
30 15983.78 16096.03 17530.06
36 17574.46 18098.36 20085.9
42 18671.14 19808.92 21875.84
48 19431.05 22036.06 23986.08
56 19737.92 22115.34 24275.72
64 19946.57 21457.32 24054.09
72 20129.7 21729.78 24167.03
80 20214.93 21594.51 24092.86
90 20194.78 21195.61 23945.93
100 20753.44 21597.26 23802.16
128 20235.24 20684.34 23476.82
192 20280.52 20431.18 23108.36
256 20410.55 20952.64 22775.63
512 20953.73 22079.18 23489.3

 

Here we see the opposite result with SystemD: Percona Server running from a bash shell shows the better throughput compared with the SystemD service. So for some reason, systemd works differently for AMD and Intel CPUs. Please let me know if you have any ideas on how to deal with the impact that systemd has on performance.

Conclusions

So there are some conclusions from these results:

  1. AMD EPYC shows a decent performance scalability; the new kernel helps to improve it
  2. systemd shows different effects on throughput for AMD and Intel CPUs
  3. With AMD the throughput declines for a high concurrent workload with 512 threads, while Intel does not show a decline.

The post AMD EPYC Performance Testing… or Don’t get on the wrong side of SystemD appeared first on Percona Database Performance Blog.

Jul
09
2018
--

Percona Live Europe 2018 Call for Papers is Now Open

Percona Live Europe Open Source Database Conference PLE 2018

Percona Live Europe Open Source Database Conference PLE 2018Announcing the opening of the Percona Live Europe Open Source Database Conference 2018 in Frankfurt, Germany call for papers. It will be open from now until August 10, 2018.

Our theme this year is
Connect. Accelerate. Innovate.

As a speaker at Percona Live Europe, you’ll have the opportunity to CONNECT with your peers—open source database experts and enthusiasts who share your commitment to improving knowledge and exchanging ideas. ACCELERATE your projects and career by presenting at the premier open source database event, a great way to build your personal and company brands. And influence the evolution of the open source software movement by demonstrating how you INNOVATE!

Community initiatives remain core to the open source ethos, and we are proud of the contribution we make with Percona Live Europe in showcasing thought leading practices in the open source database world.

With a nod to innovation, for the first time, this year we are introducing a business track to benefit those business leaders who are exploring the use of open source and are interested in learning more about its costs and benefits.

Speaking Opportunities

The Percona Live Europe Open Source Database Conference 2018 Call for Papers is open until August 10, 2018. We invite you to submit your speaking proposal for breakout, tutorial or lightning talk sessions. Classes and talks are invited for Foundation (either entry level or of general interest to all), Core (intermediate) and Masterclass (advanced) levels.

If selected, you will receive a complimentary full conference pass.

  • Breakout Session. Broadly cover a technology area using specific examples. Sessions should be either 25 minutes or 50 minutes in length (including Q&A).
  • Tutorial Session. Present a technical session that aims for a level between a training class and a conference breakout session. We encourage attendees to bring and use laptops for working on detailed and hands-on presentations. Tutorials will be three or six hours in length (including Q&A).
  • Lightning Talk. Give a five-minute presentation focusing on one key point that interests the open source community: technical, lighthearted or entertaining talks on new ideas, a successful project, a cautionary story, a quick tip or demonstration.

Topics and Tracks

We want proposals that cover the many aspects of application development using all open source databases, as well as new and interesting ways to monitor and manage database environments. Did you just embrace open source databases this year? What are the technical and business values of moving to or using open source databases? How did you convince your company to make the move? Was there tangible ROI?

Best practices and current trends, including design, application development, performance optimization, HA and clustering, cloud, containers and new technologies, as well as new and interesting ways to monitor and manage database environments—what’s holding your focus? Share your case studies, experiences and technical knowledge with an engaged audience of open source peers.

In the submission entry you will be asked to indicate which of these tracks your proposal best fits: tutorial, business needs; case studies/use cases; operations; or developer.

A few ideas

The conference committee is looking for proposals that cover the many aspects of using, deploying and managing open source databases, including:

  • Open source – Describe the technical and business values of moving to or using open source databases. How did you convince your company to make the move? Was there tangible ROI?
  • Security – All of us have experienced security challenges. Whether they are initiated by legislature (GDPR), bugs (Meltdown/Spectre), experience (external attacks) or due diligence (planning for the worst), when do you have ‘enough’ security? Are you finding that security requirements are preventing your ability to be agile?
  • Serverless, Cloud or On-Premise – The technology landscape is no longer a simple one, and mixing infrastructures has almost become the norm. Are you designing data architectures for the new landscape, and eager to share your experience? Have microservices become an important part of your plans?
  • MySQL – Do you have an opinion on what is new and exciting in MySQL? With the release of MySQL 8.0, are you using the latest features? How and why? Are they helping you solve any business issues, or making deployment of applications and websites easier, faster or more efficient? Did the new release get you to change to MySQL? What do you see as the biggest impact of the MySQL 8.0 release? Do you use MySQL in conjunction with other databases in your environment?
  • MongoDB – How has the 3.6 release improved your experience in application development or time-to-market? How are the new features making your database environment better? What is it about MongoDB 4.0 that excites you? What are your experiences with Atlas? Have you moved to it, and has it lived up to its promises? Do you use MongoDB in conjunction with other databases in your environment?
  • PostgreSQL – Why do you use PostgreSQL as opposed to other SQL options? Have you done a comparison or benchmark of PostgreSQL vs. other types of databases related to your tasks? Why and what were the results? How does PostgreSQL help you with application performance or deployment? How do you use PostgreSQL in conjunction with other databases in your environment?
  • SQL, NewSQL, NoSQL – It’s become a perennial question without an easy answer. How do databases compare, how do you choose the right technology for the job, how do you trade off between features and their benefits in comparing databases? If you have ever tried a hybrid database approach in a single application, how did that work out? How nicely does MongoDB play with MySQL in the real world? Do you have anything to say about using SQL with NoSQL databases?
  • High Availability – What choices are you making to ensure high availability? How do you find the balance between redundancy and cost? Are you using hot backups, and if so, what happened when you needed to rollback on them?
  • Scalability – When did you recognize you needed to address data scale? Did your data growth take you by surprise or were you always in control? Did it take a degradation in performance to get your management to sit up and take notice? How do you plan for scale if you can’t predict demand?
  • What the Future Holds – What do you see as the “next big thing”? What new and exciting features are going to be released? What’s in your next release? What new technologies will affect the database landscape? AI? Machine learning? Blockchain databases? Let us know about innovations you see on the way.

How to respond to the call for papers

For information on how to submit your proposal visit our call for papers page. The conference web pages will be updated throughout the next few weeks and bios, synopsis and slides will be published on those pages after the event.

Sponsorship

If you would like to obtain a sponsor pack for Percona Live Europe Open Source Database Conference 2018, you will find more information including a prospectus on our sponsorship page. You are welcome to contact me, Bronwyn Campbell, directly.

The post Percona Live Europe 2018 Call for Papers is Now Open appeared first on Percona Database Performance Blog.

Jul
06
2018
--

Percona Toolkit 3.0.11 Is Now Available

percona toolkit

percona toolkitPercona announces the release of Percona Toolkit 3.0.11 on July 6, 2018.

Percona Toolkit is a collection of advanced open source command-line tools, developed and used by the Percona technical staff, that are engineered to perform a variety of MySQL®, MongoDB® and system tasks that are too difficult or complex to perform manually. With over 1,000,000 downloads, Percona Toolkit supports Percona Server for MySQL, MySQL®, MariaDB®, Percona Server for MongoDB and MongoDB.

Percona Toolkit, like all Percona software, is free and open source. You can download packages from the website or install from official repositories.

This release includes the following changes:

New Features:

  • PT-1571: Improved hostname recognition in pt-secure-collect
  • PT-1569: Disabled --alter-foreign-keys-method=drop_swap in pt-online-schema-change
  • PT-242: (pt-stalk) Include SHOW SLAVE STATUS on MySQL 5.7 (Thanks Marcelo Altmann)

Fixed bugs:

  • PT-1570: pt-archiver fails to detect columns with the word *GENERATED* as part of the comment
  • PT-1563: pt-show-grantsfails for MySQL 5.6 producing an error which reports that an unknown column account_locked has been detected.
  • PT-1551: pt-table-checksum fails on MySQL 8.0.11
  • PT-241: (pt-stalk) Slave queries don’t run on MySQL 5.7  because the FQDN was missing (Thanks Marcelo Altmann)

Breaking changes:

Starting with this version, the queries checksum in pt-query-digest will use the full MD5 field as a CHAR(32) field instead of storing just the least significant bytes of the checksum as a BIGINT field. The reason for this change is that storing only the least significant bytes as a BIGINT was producing inconsistent results in MySQL 8 compared to MySQL 5.6+.

pt-online-schema-change in MySQL 8:

Due to a bug in MySQL 8.0+, it is not possible to use the drop_swapmethod to rebuild constraints because renaming a table will result in losing the foreign keys. You must specify a different method explicitly.

Help us improve our software quality by reporting any bugs you encounter using our bug tracking system.

The post Percona Toolkit 3.0.11 Is Now Available appeared first on Percona Database Performance Blog.

Jul
06
2018
--

Another Day, Another Data Leak

another day another data leak Exactis

another day another data leak ExactisIn the last few days, there has been information released about yet another alleged data leak, placing in jeopardy “…[the] personal information on hundreds of millions of American adults, as well as millions of businesses.” In this case, the “victim” was Exactis, for whom data collection and data security are core business functions.

Some takeaways from Exactis

Please excuse the pun! In security, we have few chances to chuckle. In fact, as a Security Architect, I sigh deeply when I read about this kind of issue. Firstly, it’s preventable. Secondly, I worry that if an organization like Exactis is not getting it right, what chance the rest of the world?

As the Wired article notes the tool https://shodan.io/ can be revealing and well worth a look. For example, you can see there are still MANY elasticSearch systems exposed to the public internet here. Why not use shodan to check what everyone else in the world can see openly on your systems ?

Securing databases

Databases in themselves do not need to be at risk, as long as you take the necessary precautions. We discussed this in this blog post that I co-authored last year.

In this latest alleged gaffe, as far as I can discern, had the setup made use of iptables or a similar feature then the breach could not have occurred.

With immaculate timing, my colleague Marco Tusa wrote a post last month on how to set up iptables for Percona XtraDB Cluster, and if you are not sure if or how that applies to your setup, it is definitely worth a read. In fact, you can access all of our security blog posts if you would like some more pointers.

Of course, security does not stop with iptables. Application developers should already be familiar with the need to avoid SQL injection, and there is a decent SQL injection prevention cheat sheet here, offered by The Open Web Application Security Project (OWASP). Even if you don’t fully understand the technical details, a cheat sheet like this might help you to ask the right questions for your application.

MySQL resources

For a more in-depth look at MySQL security, I have two talks up on YouTube. The first of these is a twenty-minute presentation on hardening MySQL and the second on web application security and why you really should review yours. You could also check out our recorded webinar Security and Encryption in the MySQL world presented by Dimitri Vanoverbeke.

MongoDB resources

Of course, security challenges are not unique to SQL databases. If you are a MongoDB user, this webinar MongoDB Security: Making things secure by default might be of interest to you. Or perhaps this one on using LDAP Authentication with MongoDB? Adamo Tonete presents both of these webinars.

For a more widely applicable view, you could try Colin Charles’ recent webinar too.

There are always consequences

As Exactis are no doubt discovering, managing the fallout from such a breach is a challenge. If you are not sure where you stand on security, or what you can do to improve your situation, then audit services such as those we offer could prove to be a valuable investment.

Finally, some of you will be lucky enough to have someone dedicated to IT security in your organizations. Next time you see them, instead of avoiding their steely stare, why not invite them for a coffee* and a chat? It could be enlightening!

*Beer or scotch is also almost always accepted too…

The post Another Day, Another Data Leak appeared first on Percona Database Performance Blog.

Jul
03
2018
--

Linux OS Tuning for MySQL Database Performance

Linux OS tuning for MySQL database performance

Linux OS tuning for MySQL database performanceIn this post we will review the most important Linux settings to adjust for performance tuning and optimization of a MySQL database server. We’ll note how some of the Linux parameter settings used OS tuning may vary according to different system types: physical, virtual or cloud. Other posts have addressed MySQL parameters, like Alexander’s blog MySQL 5.7 Performance Tuning Immediately After Installation. That post remains highly relevant for the latest versions of MySQL, 5.7 and 8.0. Here we will focus more on the Linux operating system parameters that can affect database performance.

Server and Operating System

Here are some Linux parameters that you should check and consider modifying if you need to improve database performance.

Kernel – vm.swappiness

The value represents the tendency of the kernel  to swap out memory pages. On a database server with ample amounts of RAM, we should keep this value as low as possible. The extra I/O can slow down or even render the service unresponsive. A value of 0 disables swapping completely while 1 causes the kernel to perform the minimum amount of swapping. In most cases the latter setting should be OK:

# Set the swappiness value as root
echo 1 > /proc/sys/vm/swappiness
# Alternatively, using sysctl
sysctl -w vm.swappiness=1
# Verify the change
cat /proc/sys/vm/swappiness
1
# Alternatively, using sysctl
sysctl vm.swappiness
vm.swappiness = 1

The change should be also persisted in /etc/sysctl.conf:

vm.swappiness = 1

Filesystems – XFS/ext4/ZFS
XFS

XFS is a high-performance, journaling file system designed for high scalability. It provides near native I/O performance even when the file system spans multiple storage devices.  XFS has features that make it suitable for very large file systems, supporting files up to 8EiB in size. Fast recovery, fast transactions, delayed allocation for reduced fragmentation and near raw I/O performance with DIRECT I/O.

The default options for mkfs.xfs are good for optimal speed, so the simple command:

# Use default mkfs options
mkfs.xfs /dev/target_volume

will provide best performance while ensuring data safety. Regarding mount options, the defaults should fit most cases. On some filesystems you can see a performance increase by adding the noatime mount option to the /etc/fstab.  For XFS filesystems the default atime behaviour is relatime, which has almost no overhead compared to noatime and still maintains sane atime values.  If you create an XFS file system on a LUN that has a battery backed, non-volatile cache, you can further increase the performance of the filesystem by disabling the write barrier with the mount option nobarrier. This helps you to avoid flushing data more often than necessary. If a BBU (backup battery unit) is not present, however, or you are unsure about it, leave barriers on, otherwise you may jeopardize data consistency. With this options on, an /etc/fstab file should look like the one below:

/dev/sda2              /datastore              xfs     defaults,nobarrier
/dev/sdb2              /binlog                 xfs     defaults,nobarrier

ext4

ext4 has been developed as the successor to ext3 with added performance improvements. It is a solid option that will fit most workloads. We should note here that it supports files up to 16TB in size, a smaller limit than xfs. This is something you should consider if extreme table space size/growth is a requirement. Regarding mount options, the same considerations apply. We recommend the defaults for a robust filesystem without risks to data consistency. However, if an enterprise storage controller with a BBU cache is present, the following mount options will provide the best performance:

/dev/sda2              /datastore              ext4     noatime,data=writeback,barrier=0,nobh,errors=remount-ro
/dev/sdb2              /binlog                 ext4     noatime,data=writeback,barrier=0,nobh,errors=remount-ro

Note: The data=writeback option results in only metadata being journaled, not actual file data. This has the risk of corrupting recently modified files in the event of a sudden power loss, a risk which is minimised with a presence of a BBU enabled controller. nobh only works with the data=writeback option enabled.

ZFS

ZFS is a filesystem and LVM combined enterprise storage solution with extended protection vs data corruption. There are certainly cases where the rich feature set of ZFS makes it an essential option to consider, most notably when advance volume management is a requirement. ZFS tuning for MySQL can be a complex topic and falls outside the scope of this blog. For further reference, there is a dedicated blog post on the subject by Yves Trudeau:

Disk Subsystem – I/O scheduler 

Most modern Linux distributions come with noop or deadline I/O schedulers by default, both providing better performance than the cfq and anticipatory ones. However it is always a good practice to check the scheduler for each device and if the value shown is different than noop or deadline the policy can change without rebooting the server:

# View the I/O scheduler setting. The value in square brackets shows the running scheduler
cat /sys/block/sdb/queue/scheduler
noop deadline [cfq]
# Change the setting
sudo echo noop > /sys/block/sdb/queue/scheduler

To make the change persistent, you must modify the GRUB configuration file:

# Change the line:
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash"
# to:
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash elevator=noop"

AWS Note: There are cases where the I/O scheduler has a value of none, most notably in AWS VM instance types where EBS volumes are exposed as NVMe block devices. This is because the setting has no use in modern PCIe/NVMe devices. The reason is that they have a very large internal queue and they bypass the IO scheduler altogether. The setting in this case is none and it is the optimal in such disks.

Disk Subsystem – Volume optimization

Ideally different disk volumes should be used for the OS installation, binlog, data and the redo log, if this is possible. The separation of OS and data partitions, not just logically but physically, will improve database performance. The RAID level can also have an impact: RAID-5 should be avoided as the checksum needed to ensure integrity is costly. The best performance without making compromises to redundancy is achieved by the use of an advanced controller with a battery-backed cache unit and preferably RAID-10 volumes spanned across multiple disks.

AWS Note: For further information about EBS volumes and AWS storage optimisation, Amazon has documentation at the following links:

https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/nvme-ebs-volumes.html

https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/storage-optimized-instances.html

Database settings

System Architecture – NUMA settings

Non-uniform memory access (NUMA) is a memory design where an SMP’s system processor can access its own local memory faster than non-local memory (the one assigned local to other CPUs). This may result in suboptimal database performance and potentially swapping. When the buffer pool memory allocation is larger than size of the RAM available local to the node, and the default memory allocation policy is selected, swapping occurs. A NUMA enabled server will report different node distances between CPU nodes. A uniformed one will report a single distance:

# NUMA system
numactl --hardware
available: 4 nodes (0-3)
node 0 cpus: 0 1 2 3 4 5 6 7
node 0 size: 65525 MB
node 0 free: 296 MB
node 1 cpus: 8 9 10 11 12 13 14 15
node 1 size: 65536 MB
node 1 free: 9538 MB
node 2 cpus: 16 17 18 19 20 21 22 23
node 2 size: 65536 MB
node 2 free: 12701 MB
node 3 cpus: 24 25 26 27 28 29 30 31
node 3 size: 65535 MB
node 3 free: 7166 MB
node distances:
node   0   1   2   3
  0:  10  20  20  20
  1:  20  10  20  20
  2:  20  20  10  20
  3:  20  20  20  10
# Uniformed system
numactl --hardware
available: 1 nodes (0)
node 0 cpus: 0 1 2 3 4 5 6 7
node 0 size: 64509 MB
node 0 free: 4870 MB
node distances:
node   0
  0:  10

In the case of a NUMA system, where numactl shows different distances across nodes, the MySQL variable innodb_numa_interleave should be enabled to ensure memory interleaving. Percona Server provides improved NUMA support by introducing the flush_caches variable. When enabled, it will help with allocation fairness across nodes. To determine whether or not allocation is equal across nodes, you can examine numa_maps for the mysqld process with this script:

# The perl script numa_maps.pl will report memory allocation per CPU node:
# 3595 is the pid of the mysqld process
perl numa_maps.pl < /proc/3595/numa_maps
N0        :     16010293 ( 61.07 GB)
N1        :     10465257 ( 39.92 GB)
N2        :     13036896 ( 49.73 GB)
N3        :     14508505 ( 55.35 GB)
active    :          438 (  0.00 GB)
anon      :     54018275 (206.06 GB)
dirty     :     54018275 (206.06 GB)
kernelpagesize_kB:         4680 (  0.02 GB)
mapmax    :          787 (  0.00 GB)
mapped    :         2731 (  0.01 GB)

Conclusion

In this blog post we examined a few important OS related settings and explained how they can be tuned for better database performance.

While you are here …

You might also find value in this recorded webinar Troubleshooting Best Practices: Monitoring the Production Database Without Killing Performance

 

The post Linux OS Tuning for MySQL Database Performance appeared first on Percona Database Performance Blog.

Jun
28
2018
--

What To Do When MySQL Runs Out of Memory: Troubleshooting Guide

MySQL memory troubleshooting

MySQL memory troubleshootingTroubleshooting crashes is never a fun task, especially if MySQL does not report the cause of the crash. For example, when MySQL runs out of memory. Peter Zaitsev wrote a blog post in 2012: Troubleshooting MySQL Memory Usage with a lots of useful tips. With the new versions of MySQL (5.7+) and performance_schema we have the ability to troubleshoot MySQL memory allocation much more easily.

In this blog post I will show you how to use it.

First of all, there are 3 major cases when MySQL will crash due to running out of memory:

  1. MySQL tries to allocate more memory than available because we specifically told it to do so. For example: you did not set innodb_buffer_pool_size correctly. This is very easy to fix
  2. There is some other process(es) on the server that allocates RAM. It can be the application (java, python, php), web server or even the backup (i.e. mysqldump). When the source of the problem is identified, it is straightforward to fix.
  3. Memory leaks in MySQL. This is a worst case scenario, and we need to troubleshoot.

Where to start troubleshooting MySQL memory leaks

Here is what we can start with (assuming it is a Linux server):

Part 1: Linux OS and config check
  1. Identify the crash by checking mysql error log and Linux log file (i.e. /var/log/messages or /var/log/syslog). You may see an entry saying that OOM Killer killed MySQL. Whenever MySQL has been killed by OOM “dmesg” also shows details about the circumstances surrounding it.
  2. Check the available RAM:
    • free -g
    • cat /proc/meminfo
  3. Check what applications are using RAM: “top” or “htop” (see the resident vs virtual memory)
  4. Check mysql configuration: check /etc/my.cnf or in general /etc/my* (including /etc/mysql/* and other files). MySQL may be running with the different my.cnf (run
    ps  ax| grep mysql

     )

  5. Run
    vmstat 5 5

     to see if the system is reading/writing via virtual memory and if it is swapping

  6. For non-production environments we can use other tools (like Valgrind, gdb, etc) to examine MySQL usage
Part 2:  Checks inside MySQL

Now we can check things inside MySQL to look for potential MySQL memory leaks.

MySQL allocates memory in tons of places. Especially:

  • Table cache
  • Performance_schema (run:
    show engine performance_schema status

      and look at the last line). That may be the cause for the systems with small amount of RAM, i.e. 1G or less

  • InnoDB (run
    show engine innodb status

      and check the buffer pool section, memory allocated for buffer_pool and related caches)

  • Temporary tables in RAM (find all in-memory tables by running:
    select * from information_schema.tables where engine='MEMORY'

     )

  • Prepared statements, when it is not deallocated (check the number of prepared commands via deallocate command by running show global status like ‘
    Com_prepare_sql';show global status like 'Com_dealloc_sql'

      )

The good news is: starting with MySQL 5.7 we have memory allocation in performance_schema. Here is how we can use it

  1. First, we need to enable collecting memory metrics. Run:
    UPDATE setup_instruments SET ENABLED = 'YES'
    WHERE NAME LIKE 'memory/%';
  2. Run the report from sys schema:
    select event_name, current_alloc, high_alloc
    from sys.memory_global_by_current_bytes
    where current_count > 0;
  3. Usually this will give you the place in code when memory is allocated. It is usually self-explanatory. In some cases we can search for bugs or we might need to check the MySQL source code.

For example, for the bug where memory was over-allocated in triggers (https://bugs.mysql.com/bug.php?id=86821) the select shows:

mysql> select event_name, current_alloc, high_alloc from memory_global_by_current_bytes where current_count > 0;
+--------------------------------------------------------------------------------+---------------+-------------+
| event_name                                                                     | current_alloc | high_alloc  |
+--------------------------------------------------------------------------------+---------------+-------------+
| memory/innodb/buf_buf_pool                                                     | 7.29 GiB      | 7.29 GiB    |
| memory/sql/sp_head::main_mem_root                                              | 3.21 GiB      | 3.62 GiB    |
...

The largest chunk of RAM is usually the buffer pool but ~3G in stored procedures seems to be too high.

According to the MySQL source code documentation, sp_head represents one instance of a stored program which might be of any type (stored procedure, function, trigger, event). In the above case we have a potential memory leak.

In addition we can get a total report for each higher level event if we want to see from the birds eye what is eating memory:

mysql> select  substring_index(
    ->     substring_index(event_name, '/', 2),
    ->     '/',
    ->     -1
    ->   )  as event_type,
    ->   round(sum(CURRENT_NUMBER_OF_BYTES_USED)/1024/1024, 2) as MB_CURRENTLY_USED
    -> from performance_schema.memory_summary_global_by_event_name
    -> group by event_type
    -> having MB_CURRENTLY_USED>0;
+--------------------+-------------------+
| event_type         | MB_CURRENTLY_USED |
+--------------------+-------------------+
| innodb             |              0.61 |
| memory             |              0.21 |
| performance_schema |            106.26 |
| sql                |              0.79 |
+--------------------+-------------------+
4 rows in set (0.00 sec)

I hope those simple steps can help troubleshoot MySQL crashes due to running out of memory.

Links to more resources that might be of interest

The post What To Do When MySQL Runs Out of Memory: Troubleshooting Guide appeared first on Percona Database Performance Blog.

Jun
27
2018
--

Percona Monitoring and Management 1.12.0 Is Now Available

Percona Monitoring and Management

Percona Monitoring and ManagementPMM (Percona Monitoring and Management) is a free and open-source platform for managing and monitoring MySQL and MongoDB performance. You can run PMM in your own environment for maximum security and reliability. It provides thorough time-based analysis for MySQL and MongoDB servers to ensure that your data works as efficiently as possible.

In release 1.12, we invested our efforts in the following areas:

  • Visual Explain in Query Analytics – Gain insight into MySQL’s query optimizer for your queries
  • New Dashboard – InnoDB Compression Metrics – Evaluate effectiveness of InnoDB Compression
  • New Dashboard – MySQL Command/Handler Compare – Contrast MySQL instances side by side
  • Updated Grafana to 5.1 – Fixed scrolling issues

We addressed 10 new features and improvements, and fixed 13 bugs.

Visual Explain in Query Analytics

We’re working on substantial changes to Query Analytics and the first part to roll out is something that users of Percona Toolkit may recognize – we’ve introduced a new element called Visual Explain based on pt-visual-explain.  This functionality transforms MySQL EXPLAIN output into a left-deep tree representation of a query plan, in order to mimic how the plan is represented inside MySQL.  This is of primary benefit when investigating tables that are joined in some logical way so that you can understand in what order the loops are executed by the MySQL query optimizer. In this example we are demonstrating the output of a single table lookup vs two table join:

Single Table Lookup Two Tables via INNER JOIN
SELECT DISTINCT c
FROM sbtest13
WHERE id
BETWEEN 49808
AND 49907
ORDER BY c
SELECT sbtest3.c
FROM sbtest1
INNER JOIN sbtest3
ON sbtest1.id = sbtest3.id
WHERE sbtest3.c ='long-string';

InnoDB Compression Metrics Dashboard

A great feature of MySQL’s InnoDB storage engine includes compression of data that is transparently handled by the database, saving you space on disk, while reducing the amount of I/O to disk as fewer disk blocks are required to store the same amount of data, thus allowing you to reduce your storage costs.  We’ve deployed a new dashboard that helps you understand the most important characteristics of InnoDB’s Compression.  Here’s a sample of visualizing Compression and Decompression attempts, alongside the overall Compression Success Ratio graph:

 

MySQL Command/Handler Compare Dashboard

We have introduced a new dashboard that lets you do side-by-side comparison of Command (Com_*) and Handler statistics.  A common use case would be to compare servers that share a similar workload, for example across MySQL instances in a pool of replicated slaves.  In this example I am comparing two servers under identical sysbench load, but exhibiting slightly different performance characteristics:

The number of servers you can select for comparison is unbounded, but depending on the screen resolution you might want to limit to 3 at a time for a 1080 screen size.

New Features & Improvements

  • PMM-2519: Display Visual Explain in Query Analytics
  • PMM-2019: Add new Dashboard InnoDB Compression metrics
  • PMM-2154: Add new Dashboard Compare Commands and Handler statistics
  • PMM-2530: Add timeout flags to mongodb_exporter (thank you unguiculus for your contribution!)
  • PMM-2569: Update the MySQL Golang driver for MySQL 8 compatibility
  • PMM-2561: Update to Grafana 5.1.3
  • PMM-2465: Improve pmm-admin debug output
  • PMM-2520: Explain Missing Charts from MySQL Dashboards
  • PMM-2119: Improve Query Analytics messaging when Host = All is passed
  • PMM-1956: Implement connection checking in mongodb_exporter

Bug Fixes

  • PMM-1704: Unable to connect to AtlasDB MongoDB
  • PMM-1950: pmm-admin (mongodb:metrics) doesn’t work well with SSL secured mongodb server
  • PMM-2134: rds_exporter exports memory in Kb with node_exporter labels which are in bytes
  • PMM-2157: Cannot connect to MongoDB using URI style
  • PMM-2175: Grafana singlestat doesn’t use consistent colour when unit is of type Time
  • PMM-2474: Data resolution on Dashboards became 15sec interval instead of 1sec
  • PMM-2581: Improve Travis CI tests by addressing pmm-admin check-network Time Drift
  • PMM-2582: Unable to scroll on “_PMM Add Instance” page when many RDS instances exist in an AWS account
  • PMM-2596: Set fixed height for panel content in PMM Add Instances
  • PMM-2600: InnoDB Checkpoint Age does not show data for MySQL
  • PMM-2620: Fix balancerIsEnabled & balancerChunksBalanced values
  • PMM-2634: pmm-admin cannot create user for MySQL 8
  • PMM-2635: Improve error message while adding metrics beyond “exit status 1”

Known Issues

  • PMM-2639: mysql:metrics does not work on Ubuntu 18.04 – We will address this in a subsequent release

How to get PMM Server

PMM is available for installation using three methods:

The post Percona Monitoring and Management 1.12.0 Is Now Available appeared first on Percona Database Performance Blog.

Jun
26
2018
--

Webinar 6/27: MySQL Troubleshooting Best Practices: Monitoring the Production Database Without Killing Performance

performance troubleshooting MySQL monitoring tools

performance troubleshooting MySQL monitoring toolsPlease join Percona’s Principal Support Escalation Specialist Sveta Smirnova as she presents Troubleshooting Best Practices: Monitoring the Production Database Without Killing Performance on Wednesday, June 27th at 11:00 AM PDT (UTC-7) / 2:00 PM EDT (UTC-4).

 

During the MySQL Troubleshooting webinar series, I covered many monitoring and logging tools such as:

  • General, slow, audit, binary, error log files
  • Performance Schema
  • Information Schema
  • System variables
  • Linux utilities
  • InnoDB monitors
  • PMM

However, I did not spend much time on the impact these instruments have on overall MySQL performance. And they do have an impact.

And this is the conflict many people face. MySQL Server users try exploring these monitoring instruments, see that they slow down their installations, and turn them off. This is unfortunate. If the instrument that can help you resolve a problem is OFF, you won’t have good and necessary information to help understand when, how and why the issue occurred. In the best case, you’ll re-enable instrumentation and wait for the next disaster occurrence. In the worst case, you try various fix options without any real knowledge if they solve the problem or not.

This is why it is important to understand the impact monitoring tools have on your database, and therefore how to minimize it.

Understanding and controlling the impact of MySQL monitoring tools

In this webinar, I cover why certain monitoring tools affect performance, and how to minimize the impact without turning the instrument off. You will learn how to monitor safely and effectively.

Register Now

 

Sveta Smirnova

Principal Support Escalation Specialist

Sveta joined Percona in 2015. Her main professional interests are problem-solving, working with tricky issues, bugs, finding patterns that can quickly solve typical issues and teaching others how to deal with MySQL issues, bugs and gotchas effectively. Before joining Percona, Sveta worked as Support Engineer in MySQL Bugs Analysis Support Group in MySQL AB-Sun-Oracle. She is the author of book “MySQL Troubleshooting” and JSON UDF functions for MySQL.

The post Webinar 6/27: MySQL Troubleshooting Best Practices: Monitoring the Production Database Without Killing Performance appeared first on Percona Database Performance Blog.

Jun
25
2018
--

Running Percona XtraDB Cluster in Kubernetes/OpenShift

Diagram of Percona XtraDB Cluster / MySQL running in Kubernetes Open Shift

Kubernetes, and its most popular distribution OpenShift, receives a lot of interest as a container orchestration platform. However, databases remain a foreign entity, primarily because of their stateful nature, since container orchestration systems prefer stateless applications. That said, there has been good progress in support for StatefulSet applications and persistent storage, to the extent that it might be already comfortable to have a production database instance running in Kubernetes. With this in mind, we’ve been looking at running Percona XtraDB Cluster in Kubernetes/OpenShift.

While there are already many examples on the Internet of how to start a single MySQL instance in Kubernetes, for serious usage we need to provide:

  • High Availability: how can we guarantee availability when an instance (or Pod in Kubernetes terminology) crashes or becomes unresponsive?
  • Persistent storage: we do not want to lose our data in case of instance failure
  • Backup and recovery
  • Traffic routing: in the case of multiple instances, how do we direct an application to the correct one
  • Monitoring

Percona XtraDB Cluster in Kubernetes/OpenShift

Schematically it looks like this:


Percona XtraDB Cluster in Kubernetes/OpenShift a possible configuration for a resilient solution

The picture highlights the components we are going to use

Running this in Kubernetes assumes a high degree of automation and minimal manual intervention.

We provide our proof of concept in this project: https://github.com/Percona-Lab/percona-openshift. Please treat it like a source for ideas and as an alpha-quality project, in no way it is production ready.

Details

In our implementation we rely on Helm, the package manager for Kubernetes.  Unfortunately OpenShift does not officially support Helm out of the box, but there is a guide from RedHat on how to make it work.

In the clustering setup, it is quite typical to use a service discovery software like Zookeeper, etcd or Consul. It may become necessary for our Percona XtraDB Cluster deployment, but for now, to simplify deployment, we are going to use the DNS service discovery mechanism provided by Kubernetes. It should be enough for our needs.

We also expect the Kubernetes deployment to provide Dynamic Storage Provisioning. The major cloud providers (like Google Cloud, Microsoft Azure or Amazon Cloud) should have it. Also, it might not be easy to have Dynamic Storage Provisioning for on-premise deployments. You may need to setup GlusterFS or Ceph to provide Dynamic Storage Provisioning.

The challenge with a distributed file system is how many copies of data you will end up having. Percona XtraDB Cluster by itself has three copies, and GlusterFS will also require at least two copies of the data, so in the end we will have six copies of the data. This can’t be good for write intensive applications, but it’s also not good from the capacity standpoint.

One possible approach is to have local data copies for Percona XtraDB Cluster deployments. This will provide better performance and less impact on the network, but in the case of a big dataset (100GB+ ) the node failure will require SST with a big impact on the cluster and network. So the individual solution should be tailored for your workload and your requirements.

Now, as we have a basic setup working, it would be good to understand the performance impact of running Percona XtraDB Cluster in Kubernetes.  Is the network and storage overhead acceptable or it is too big? We plan to look into this in the future.

Once again, our project is located at https://github.com/Percona-Lab/percona-openshift, we are looking for your feedback and for your experience of running databases in Kubernetes/OpenShift.

Before you leave …

Percona XtraDB Cluster

If this article has interested you and you would like to know more about Percona XtraDB Cluster, you might enjoy our recent series of webinar tutorials that introduce this software and how to use it.

The post Running Percona XtraDB Cluster in Kubernetes/OpenShift appeared first on Percona Database Performance Blog.

Powered by WordPress | Theme: Aeros 2.0 by TheBuckmaker.com