Personal blog of Yzmir Ramirez

Percona announces the GA release of Percona Server for MySQL 5.7.22-22 on on May 31, 2018. Download the latest version from the Percona web site or the Percona Software Repositories. You can also run Docker containers from the images in the Docker Hub repository.

Based on MySQL 5.7.22, including all the bug fixes in it, Percona Server for MySQL 5.7.22-22 is the current GA release in the Percona Server for MySQL 5.7 series. Percona provides completely open-source and free software.

New Features:

• A new --encrypt-tmp-files option turns on encryption for the temporary files which Percona Server may create on disk for filesort, binary log transactional caches and Group Replication caches.

Bugs Fixed:

• Executing the SHOW GLOBAL STATUS expression could cause “data drift” on global status variables in case of a query rollback: the variable, being by its nature a counter and allowing only an increase, could return to its previous value. Bug fixed #3951 (upstream #90351).
• NUMA support was improved in Percona Server, reverting upstream implementation back to the original one,due to upstream variant being less effective in memory allocation. Now  innodb_numa_interleave variable not only enables NUMA interleave memory policy for the InnoDB buffer pool allocation, but forces NUMA interleaved allocation at the buffer pool initialization time. Bug fixed #3967.
• audit_log_include_accounts variable did not take effect if placed in my.cnf configuration file, while still working as intended if set dynamically. Bug fixed #3867.
• A key_block_size value was set automatically by the Improved MEMORY Storage Engine, which resulted in warnings when changing the engine type to InnoDB, and constantly growing key_block_size during alter operations. Bugs fixed #3936, #3940, and #3943.
• Fixes were introduced to remove GCC 8 compilation warnings for the Percona Server build. Bug fixed #3950.
• An InnoDB Memcached Plugin code clean-up was backported from MySQL 8.0. Bug fixed  #4506.
• Percona Server could not be built with -DWITH_LZ4=system option on Ubuntu 14.04 (Trusty) because of too old LZ4 packages. Bug fixed #3842.
• A regression brought during TokuDB code clean-up in 5.7.21-21 was causing assertion in cases when the FT layer returns an error during an alter table operation. Bug fixed #4294.

MyRocks Changes and fixes:

• UPDATE statements were returning incorrect results because of not making a full table scan on tables with unique secondary index. Bug fixed #4495 (upstream facebook/mysql-5.6#830).

Other Bugs Fixed:

• #4451 “Implement better compression algo testing”
• #4469 “variable use out of scope bug in get_last_key test detected by ASAN in clang 6”
• #4470 “the cachetable-simple-pin-nonblocking-cheap test occasionally fails due to a locking conflict with the cachetable evictor”
• #4488 “-Werror is always disabled for innodb_memcached”
• #1114 “Assertion `inited == INDEX’ failed”
• #1130 “RBR Replication with concurrent XA in READ-COMMITTED takes supremum pseudo-records and breaks replication”

Find the release notes for Percona Server for MySQL 5.7.22-22 in our online documentation. Report bugs in the Jira bug tracker.

The post Percona Server for MySQL 5.7.22-22 Is Now Available appeared first on Percona Database Performance Blog.

If you’ve ever been in a pointless meeting at work, odds are you’ve spent part of the time responding to messages or just putzing around on the Internet — but Klaxoon hopes to convert that into something a bit more productive with more interactive meetings.

The French startup today said it’s raised $50 million in a new financing round led by Idinvest Partners, with early round investors BPI, Sofiouest, Arkea and White Star Capital Fund also participating. The company offers a suite of tools designed to make those meetings more engaging and generally just cut down on useless meetings with a room of bored and generally unengaged people that might be better off working away at their desk or even taking other meetings. The company has raised about $55.6 million in total.

The whole point of Klaxoon is to make meetings more engaging, and there are a couple ways to do that. The obvious point is to translate what some classrooms are doing in the form of making the whole session more engaging with the use of connected devices. You might actually remember those annoying clickers in classrooms used to answer multiple choice questions throughout a session, but it is at least one way to engage people in a room — and offering a more robust way of doing that may be something that helps making the session as a whole more productive.

Klaxoon also offers other tools like an interactive whiteboard (remember Smartboards, also in classrooms?) as well as a closed networks for meeting participants that aims to be air-gapped from a broader network so those employees can conduct a meeting in private or if the room isn’t available. All this is wrapped together with a set of analytics to help employees — or managers — better conduct meetings and generally be more productive. All this is going to be more important going forward as workplaces become more distributed, and it may be tempting to just have a virtual meeting on one screen while either working on a different one — or just messing around on the Internet.

Of course, lame meetings are a known issue — especially within larger companies. So there are multiple interpretations of ways to try to fix that problem, including Worklytics — a company that came out of Y Combinator earlier this year — that are trying to make teams more efficient in general. The idea is that if you are able to reduce the time spent in meetings that aren’t really productive, that’ll increase the output of a team in general. The goal is not to monitor teams closely, but just find ways to encourage them to spend their time more wisely. Creating a better set of productivity tools inside those meetings is one approach, and one Klaxoon seems to hope plays out.

A data lake is “…a method of storing data within a system or repository, in its natural format, that facilitates the collocation of data in various schemata and structural forms…”¹. Many companies find value in using a data lake but aren’t clear that they need to properly plan for it and maintain it in order to prevent issues.

The idea of a data lake rose from the need to store data in a raw format that is accessible to a variety of applications and authorized users. Hadoop is often used to query the data, and the necessary structures for querying are created through the query tool (schema on read) rather than as part of the data design (schema on write). There are other tools available for analysis, and many cloud providers are actively developing additional options for creating and managing your data lake. The cloud is often viewed as an ideal place for your data lake since it is inherently elastic and can expand to meet the needs of your data.

Data Lake or Data Swamp?

One of the key components of a functioning data lake is the continuing inflow and egress of data. Some data must be kept indefinitely but some can be archived or deleted after a defined period of time. Failure to remove stale data can result in a data swamp, where the out of date data is taking up valuable and costly space and may be causing queries to take longer to complete. This is one of the first issues that companies encounter in maintaining their data lake. Often, people view the data lake as a “final resting place” for data, but it really should be used for data that is accessed often, or at least occasionally.

A natural spring-fed lake can turn into a swamp due to a variety of factors. If fresh water is not allowed to flow into the lake, this can cause stagnation, meaning that plants and animals that previously were not able to be supported by the lake take hold. Similarly, if water cannot exit the lake at some point, the borders will be breached, and the surrounding land will be inundated. Both of these conditions can cause a once pristine lake to turn into a fetid and undesirable swamp. If data is no longer being added to your data lake, the results will become dated and eventually unreliable. Also, if data is always being added to the lake but is not accessed on a regular basis, this can lead to unrestricted growth of your data lake, with no real plan for how the data will be used. This can become an expensive “cold storage” facility that is likely more expensive than archived storage.

If bad or undesirable items, like old cars or garbage, are thrown into a lake, this can damage the ecosystem, causing unwanted reactions. In a data lake, this is akin to simply throwing data into the data lake with no real rules or rationale. While the data is saved, it may not be useful and can cause negative consequences across the whole environment since it is consuming space and may slow response times. Even though a basic concept of a data lake is that the data does not need to conform to a predefined structure, like you would see with a relational database, it is important that some rules and guidelines exist regarding the type and quality of data that is included in the lake. In the absence of some guidelines, it becomes difficult to access the relevant data for your needs. Proper definition and tagging of content help to ensure that the correct data is accessible and available when needed.

Unrestricted Growth Consequences

Many people have a junk drawer somewhere in their house; a drawer that is filled with old receipts, used tickets, theater programs, and the like. Some of this may be stored for sentimental reasons, but a lot of it is put into this drawer since it was a convenient dropping place for things. Similarly, if we look to the data lake as the “junk drawer” for our company, it is guaranteed to be bigger and more expensive than it truly needs to be.

It is important that the data that is stored in your data lake has a current or expected purpose. While you may not have a current use for some data, it can be helpful to keep it around in case a need arises. An example of this is in the area of machine learning. Providing more ancillary data enables better decisions since it provides a deeper view into the decision process. Therefore, maintaining some data that may not have a specific and current need can be helpful. However, there are cases where maintaining a huge volume of data can be counterproductive. Consider temperature information delivered from a switch. If the temperature reaches a specific threshold, the switch should be shut down. Reporting on the temperature in an immediate and timely manner is important to make an informed decision, but stable temperature data from days, week, or months ago could be summarized and stored in a more efficient manner. The granular details can then be purged from the lake.

So, where is the balance? If you keep all the data, it can make your data lake unwieldy and costly. If you only keep data that has a specific current purpose, you may be impairing your future plans. Obviously, the key is to monitor your access and use of the data frequently, and purge or archive some of the data that is not being regularly used.

Uncontrolled Access Concerns

Since much of the data in your data lake is company confidential, it is imperative that access to that data be controlled. The fact that the data in the lake is stored in its raw format means that it is more difficult to control access. The structures of a relational database provide some of the basis for access control, allowing us to limit who has access to specific queries, tables, fields, schemas, databases, and other objects. In the absence of these structures, controlling access requires more finesse. Determining who has access to what parts of the data in the lake must be handled, as well as isolating the data within your own network environment. Many of these restrictions may already be in place in your current environment, but they should be reviewed before being relied on fully, since the data lake may store information that was previously unavailable to some users. Access should be regularly reviewed to identify potential rogue activities. Encryption options also exist to further secure the data from unwanted access, and file system security can be used to limit access. All of these components must be considered, implemented, and reviewed to ensure that the data is secure.

User Considerations

In a relational database, the data structure inherently determines some of the consistencies and format of the data. This enables users to easily query the data and be assured that they are returning valid results. The lack of such structures in the data lake means that users must be more highly skilled at data manipulation. Having users with less skill accessing the data is possible, but it may not provide the best results. A data scientist is better positioned to access and query the complete data set. Obviously, users with a higher skill set are rare and cost more to hire, but the return may be worth it in the long run.

So What Do I Do Now?

This is an area where there are no hard and fast rules. Each company must develop and implement processes and procedures that make sense for their individual needs. Only with a plan for monitoring inputs, outputs, access patterns, and the like are you able to make a solid determination for your company’s needs. Percona can help to determine a plan for reporting usage, assess security settings, and more. As you are using the data in your data lake, we can also provide guidance regarding tools used to access the data.

¹ Wikipedia, May 22, 2018

The post Don’t Drown in your Data Lake appeared first on Percona Database Performance Blog.

Amazon QuickSight, the company’s business intelligence tool for AWS, launched back in 2015, but it’s hard to say how much impact the service has made in the highly competitive BI market. The company has far from given up on this project, though, and today, it’s introducing a new pay-per-session pricing plan for access to QuickSight dashboards that is surely meant to give it a bit of a lift in a market where Tableau and Microsoft’s Power BI have captured much of the mindshare.

Under the new pricing plan, creating and publishing dashboards will stay cost $18 per user and month. For readers, though, who only need to have access to these dashboards, AWS now offers a very simple option: they will now pay $0.30 per session up to a maximum of $5 per month and user. Under this scheme, a session is defined as the first 30 minutes from login.

Previously, AWS offered two tiers of QuickSight plans: a $9 per user/month standard plan and a $24/user/month enterprise edition with support for Active Directory and encryption at rest.

That $9/user/month is still available and probably still makes sense for smaller companies where those who build dashboards and consume them are often the same person. The new pricing plan replaces the existing enterprise edition.

QuickSight already significantly undercuts the pricing of services like Tableau and others, though we’re also talking about a somewhat more limited feature set. This new pay-per-session offering only widens the pricing gap.

“With highly scalable object storage in Amazon Simple Storage Service (Amazon S3), data warehousing at one-tenth the cost of traditional solutions in Amazon Redshift, and serverless analytics offered by Amazon Athena, customers are moving data into AWS at an unprecedented pace,” said Dorothy Nicholls, Vice President of Amazon QuickSight at AWS, in a canned comment. “What’s changed is that virtually all knowledge workers want easy access to that data and the insights that can be derived. It’s been cost-prohibitive to enable that access for entire companies until the Amazon QuickSight pay-per-session pricing — this is a game-changer in terms of information and analytics access.”

Current QuickSight users include the NFL, Siemens, Volvo and AutoTrader.

In this third and final post of the series, we look at how to configure transport encryption on a deployed MongoDB replica set. Security vulnerabilities can arise when internal personnel have legitimate access to the private network, but should not have access to the data. Encrypting intra-node traffic ensures that no one can “sniff” sensitive data on the network.

In part 1 we described MongoDB replica sets and how they work.
In part 2 we provided a step-by-step guide to deploy a simple 3-node replica set, including information on replica set configuration.

Enable Role-Based Access Control

In order for the encryption to be used in our replica set, we need first to activate Role-Based Access Control (RBAC). By default, a MongoDB installation permits anyone to connect and see the data, as in the sample deployment we created in part 2. Having RBAC enabled is mandatory for encryption.

RBAC governs access to a MongoDB system. Users are created and assigned privileges to access specific resources, such as databases and collections. Likewise, for carrying out administrative tasks, users need to be created with specific grants. Once activated, every user must authenticate themselves in order to access MongoDB.

Prior to activating RBAC, let’s create an administrative user. We’ll connect to the PRIMARY member and do the following:

rs-test:PRIMARY> use admin
switched to db admin
rs-test:PRIMARY> db.createUser({user: 'admin', pwd: 'secret', roles:['root']})
Successfully added user: { "user" : "admin", "roles" : [ "root" ] }

Let’s activate the RBAC in the configuration file /etc/mongod.conf on each node

security:
      authorization: enabled

and restart the daemon

sudo service mongod restart

Now to connect to MongoDB we issue the following command:

mongo -u admin -p secret --authenticationDatabase "admin"

Certificates

MongoDB supports X.509 certificate authentication for use with a secure TLS/SSL connection. The members can use X.509 certificates to verify their membership of the replica set.

In order to use encryption, we need to create certificates on all the nodes and have a certification authority (CA) that signs them. Since having a certification authority can be quite costly, we decide to use self-signed certificates. For our purposes, this solution ensures encryption and has no cost. Using a public CA is not necessary inside a private infrastructure.

To proceed with certificate generation we need to have openssl installed on our system and certificates need to satisfy these requirements:

• any certificate needs to be signed by the same CA
• the common name (CN) required during the certificate creation must correspond to the hostname of the host
• any other field requested in the certificate creation should be a non-empty value and, hopefully, should reflect our organization details
• it is also very important that all the fields, except the CN, should match those from the certificates for the other cluster members

The following guide describes all the steps to configure internal X.509 certificate-based encryption.

1 – Connect to one of the hosts and generate a new private key using openssl

openssl genrsa -out mongoCA.key -aes256 8192

We have created a new 8192 bit private key and saved it in the file mongoCA.key
Remember to enter a strong passphrase when requested.

2 – Sign a new CA certificate

Now we are going to create our “fake” local certification authority that we’ll use later to sign each node certificate.

During certificate creation, some fields must be filled out. We could choose these randomly but they should correspond to our organization’s details.

root@psmdb1:~# openssl req -x509 -new -extensions v3_ca -key mongoCA.key -days 365 -out
    mongoCA.crt
    Enter pass phrase for mongoCA.key:
    You are about to be asked to enter information that will be incorporated
    into your certificate request.
    What you are about to enter is what is called a Distinguished Name or a DN. There are quite a few fields but you can leave some blank
    For some fields there will be a default value,
    If you enter '.', the field will be left blank.
    -----
    Country Name (2 letter code) [AU]:US
    State or Province Name (full name) [Some-State]:California
    Locality Name (eg, city) []:San Francisco
    Organization Name (eg, company) [Internet Widgits Pty Ltd]:My Company Ltd
    Organizational Unit Name (eg, section) []:DBA
    Common Name (e.g. server FQDN or YOUR name) []:psmdb
    Email Address []:corrado@mycompany.com

3 – Issue self-signed certificates for all the nodes

For each node, we need to generate a certificate request and sign it using the CA certificate we created in the previous step.

Remember: fill out all the fields requested the same for each host, but remember to fill out a different common name (CN) that must correspond to the hostname.

For the first node issue the following commands.

openssl req -new -nodes -newkey rsa:4096 -keyout psmdb1.key -out psmdb1.csr
openssl x509 -CA mongoCA.crt -CAkey mongoCA.key -CAcreateserial -req -days 365 -in psmdb1.csr -out psmdb1.crt
cat psmdb1.key psmdb1.crt > psmdb1.pem

for the second node

openssl req -new -nodes -newkey rsa:4096 -keyout psmdb2.key -out psmdb2.csr
openssl x509 -CA mongoCA.crt -CAkey mongoCA.key -CAcreateserial -req -days 365 -in psmdb2.csr -out psmdb2.crt
cat psmdb2.key psmdb2.crt > psmdb2.pem

and for the third node

openssl req -new -nodes -newkey rsa:4096 -keyout psmdb3.key -out psmdb3.csr
openssl x509 -CA mongoCA.crt -CAkey mongoCA.key -CAcreateserial -req -days 365 -in psmdb3.csr -out psmdb3.crt
cat psmdb3.key psmdb3.crt > psmdb3.pem

4 – Place the files

We could execute all of the commands in the previous step on the same host, but now we need to copy the generated files to the proper nodes:

• Copy to each node the CA certifcate file: mongoCA.crt
• Copy each self signed certifcate <hostname>.pem into the relative member
• Create on each member a directory that only the MongoDB user can read, and copy both files there

sudo mkdir -p /etc/mongodb/ssl
sudo chmod 700 /etc/mongodb/ssl
sudo chown -R mongod:mongod /etc/mongodb
sudo cp psmdb1.pem /etc/mongodb/ssl
sudo cp mongoCA.crt /etc/mongodb/ssl

Do the same on each host.

5 – Configure mongod

Finally, we need to instruct mongod about the certificates to enable the encryption.

Change the configuration file /etc/mongod.conf on each host adding the following rows:

net:
   port: 27017
   ssl:
      mode: requireSSL
      PEMKeyFile: /etc/mongodb/ssl/psmdb1.pem
      CAFile: /etc/mongodb/ssl/mongoCA.crt
      clusterFile: /etc/mongodb/ssl/psmdb1.pem
   security:
      authorization: enabled
      clusterAuthMode: x509

Restart the daemon

sudo service mongodb restart

Make sure to put the proper file names on each host (psmdb2.pem on psmdb2 host and so on)

Now, as long as we have made no mistakes, we have a properly configured replica set that is using encrypted connections.

Issue the following command to connect on node psmdb1:

mongo admin --ssl --sslCAFile /etc/mongodb/ssl/mongoCA.crt
--sslPEMKeyFile /etc/mongodb/ssl/psmdb1.pem
-u admin -p secret --host psmdb1

Access the first two articles in this series

• Part 1: Introduces basic replica set concepts, how it works and what its main features
• Part 2:  Provides a step-by-step guide to configure a three-node replica set

The post MongoDB: deploy a replica set with transport encryption (part 3/3) appeared first on Percona Database Performance Blog.

The Europas Unconference & Awards is back on 3 July in London and we’re excited to announce more speakers and panel sessions as the event takes shape. Crypto and Blockchain will be a major theme this year, and we’re bringing together many of the key players. TechCrunch is once again the key media partner, and if you attend The Europas you’ll be first in the queue to get offers for TC events and Disrupt in Europe later in the year.

You can also potentially get your ticket for free just by sharing your own ticket link with friends and followers. See below for the details and instructions.

To recap, we’re jumping straight into our popular breakout sessions where you’ll get up close and personal with some of Europe’s leading investors, founders and thought leaders.

The Unconference is focused into zones including AI, Fintech, Mobility, Startups, Society, and Enterprise and Crypto / Blockchain.

Our Crypto HQ will feature two tracks of panels, one focused on investing and the other on how blockchain is disrupting everything from financial services, to gaming, to social impact to art.

We’ve lined up some of the leading blockchain VCs to talk about what trends and projects excite them most, including Outlier Ventures’ Jamie Burke, KR1’s George McDonaugh, blockchain angel Nancy Fenchay, Fabric Ventures’ Richard Muirhead and Michael Jackson of Mangrove Capital Partners.

Thinking of an ICO vs crowdfunding? Join Michael Jackson on how ICOs are disrupting venture capital and Ali Ganjavian, co-founder of Studio Banana, the creators of longtime Kickstarter darling OstrichPillow to understand the ins and outs of both.

We’ve also lined up a panel to discuss the process of an ICO – what do you need to consider, the highs, the lows, the timing and the importance of community. Linda Wang, founder and CEO of Lending Block, which recently raised $10 million in an April ICO, joins us.

We are thrilled to announce that Civil, the decentralised marketplace for sustainable journalism, will be joining to talk about the rise of fake news and Verisart’s Robert Norton will share his views on stamping out fraud in the art world with blockchain. Min Teo of ConsenSys will discuss blockchain and social impact and Jeremy Millar, head of Consensys UK, will speak on Smart Contracts.

Our Pathfounders Startup Zone is focused purely on startups. Our popular Meet the Press panel is back where some of tech’s finest reporters will tell you what makes a great tech story, and how to pitch (and NOT pitch them). For a start, TechCrunch’s Steve O’Hear and Quartz’s Joon Ian Wong are joining.

You’ll also hear from angels and investors including Seedcamp’s Carlos Eduardo Espinal; Eileen Burbidge of Passion Capital; Accel Partners’ Andrei Brasoveanu; Jeremy Yap; Candice Lo of Blossom Capital; Scott Sage of Crane Venture Partners; Tugce Ergul of Angel Labs; Stéphanie Hospital of OneRagtime; Connect Ventures’ Sitar Teli and Jason Ball of Qualcomm Ventures.

Sound great? You can grab your ticket here.

All you need to do is share your personal ticket link. Your friends get 15% off, and you get 15% off again when they buy.

The more your friends buy, the more your ticket cost goes down, all the way to free!

The Public Voting in the awards ends 11 June 2018 11:59: https://theeuropas.polldaddy.com/s/theeuropas2018

We’re still looking for sponsor partners to support these editorially curated panels.

Please get in touch with Petra@theeuropas.com for more details.

SPEAKERS SO FAR:

Jamie Burke, Outlier Ventures

Jeremy Millar, ConsenSys

Linda Wang, Lending Block

Robert Norton, Verisart

George McDonaugh, KR1

Eileen Burbidge, Passion Capital

Carlos Eduardo Espinal, Seedcamp

Sitar Teli, Connect Ventures

Michael Jackson, Mangrove Capital Partners

Min Teo, ConsenSys

Steve O’Hear, TechCrunch

Joon Ian Wong, Quartz

Richard Muirhead, Fabric Ventures

Nancy Fechnay, Blockchain Technologist + Angel

Candice Lo, Blossom Capital

Scott Sage, Crane Venture Partners

Andrei Brasoveanu, Accel

Tina Baker, Jag Shaw Baker

Jeremy Yap

Candice Lo, Blossom Capital

Tugce Ergul, Angel Labs

Stéphanie Hospital, OneRagtime

Jason Ball, Qualcomm Ventures

The Europas Awards
The Europas Awards are based on voting by expert judges and the industry itself. But key to the daytime is all the speakers and invited guests. There’s no “off-limits speaker room” at The Europas, so attendees can mingle easily with VIPs and speakers.

Vote for your Favourite Startups

Public Voting is still humming along. Please remember to vote for your favourite startups!

Awards by category:

Hottest Media/Entertainment Startup

Hottest E-commerce/Retail Startup

Hottest Education Startup

Hottest Startup Accelerator

Hottest Marketing/AdTech Startup

Hottest Games Startup

Hottest Mobile Startup

Hottest FinTech Startup

Hottest Enterprise, SaaS or B2B Startup

Hottest Hardware Startup

Hottest Platform Economy / Marketplace

Hottest Health Startup

Hottest Cyber Security Startup

Hottest Travel Startup

Hottest Internet of Things Startup

Hottest Technology Innovation

Hottest FashionTech Startup

Hottest Tech For Good

Hottest A.I. Startup

Fastest Rising Startup Of The Year

Hottest GreenTech Startup of The Year

Hottest Startup Founders

Hottest CEO of the Year

Best Angel/Seed Investor of the Year

Hottest VC Investor of the Year

Hottest Blockchain/Crypto Startup Founder(s)

Hottest Blockchain Protocol Project

Hottest Blockchain DApp

Hottest Corporate Blockchain Project

Hottest Blockchain Investor

Hottest Blockchain ICO (Europe)

Hottest Financial Crypto Project

Hottest Blockchain for Good Project

Hottest Blockchain Identity Project

Hall Of Fame Award – Awarded to a long-term player in Europe

The Europas Grand Prix Award (to be decided from winners)

The Awards celebrates the most forward thinking and innovative tech & blockchain startups across over some 30+ categories.

Startups can apply for an award or be nominated by anyone, including our judges. It is free to enter or be nominated.

Instead of thousands and thousands of people, think of a great summer event with 1,000 of the most interesting and useful people in the industry, including key investors and leading entrepreneurs.

• No secret VIP rooms, which means you get to interact with the Speakers

• Key Founders and investors speaking; featured attendees invited to just network

• Expert speeches, discussions, and Q&A directly from the main stage

• Intimate “breakout” sessions with key players on vertical topics

• The opportunity to meet almost everyone in those small groups, super-charging your networking

• Journalists from major tech titles, newspapers and business broadcasters

• A parallel Founders-only track geared towards fund-raising and hyper-networking

• A stunning awards dinner and party which honors both the hottest startups and the leading lights in the European startup scene

• All on one day to maximise your time in London. And it’s sunny (probably)!

That’s just the beginning. There’s more to come…

Nvidia launched a monster box yesterday called the HGX-2, and it’s the stuff that geek dreams are made of. It’s a cloud server that is purported to be so powerful it combines high performance computing with artificial intelligence requirements in one exceptionally compelling package.

You know you want to know the specs, so let’s get to it: It starts with 16x NVIDIA Tesla V100 GPUs. That’s good for 2 petaFLOPS for AI with low precision, 250 teraFLOPS
for medium precision and 125 teraFLOPS for those times when you need the highest precision. It comes standard with a 1/2 a terabyte of memory and 12 Nvidia NVSwitches, which enable GPU to GPU communications at 300 GB per second. They have doubled the capacity from the HGX-1 released last year.

Chart: Nvidia

Paresh Kharya, group product marketing manager for Nvidia’s Tesla data center products says this communication speed enables them to treat the GPUs essentially as a one giant, single GPU. “And what that allows [developers] to do is not just access that massive compute power, but also access that half a terabyte of GPU memory as a single memory block in their programs,” he explained.

Graphic: Nvidia

Unfortunately you won’t be able to buy one of these boxes. In fact, Nvidia is distributing them strictly to resellers, who will likely package these babies up and sell them to hyperscale datacenters and cloud providers. The beauty of this approach for cloud resellers is that when they buy it, they have the entire range of precision in a single box, Kharya said

“The benefit of the unified platform is as companies and cloud providers are building out their infrastructure, they can standardize on a single unified architecture that supports the entire range of high performance workloads. So whether it’s AI, or whether it’s high performance simulations the entire range of workloads is now possible in just a single platform,”Kharya explained.

He points out this is particularly important in large scale datacenters. “In hyperscale companies or cloud providers, the main benefit that they’re providing is the economies of scale. If they can standardize on the fewest possible architectures, they can really maximize the operational efficiency. And what HGX allows them to do is to standardize on that single unified platform,” he added.

As for developers, they can write programs that take advantage of the underlying technologies and program in the exact level of precision they require from a single box.

The HGX-2 powered servers will be available later this year from partner resellers including Lenovo, QCT, Supermicro and Wiwynn.

Percona announces the release of Percona Server for MySQL 5.6.40-84.0 on May 30, 2018 (downloads are available here and from the Percona Software Repositories). Based on MySQL 5.6.40, including all the bug fixes in it, Percona Server for MySQL 5.6.40-84.0 is now the current GA release in the 5.6 series. All of Percona’s software is open-source and free.

New Features

• A new string variable version_suffix allows to change suffix for the Percona Server version string returned by the read-only version variable. This allows to append the version number for the server with a custom suffix to reflect some build or configuration specifics. Also version_comment (default value of which is taken from the CMake COMPILATION_COMMENT option) is converted from a global read-only to a global read-write variable and thereby it is now cutomizable.
• Query response time plugin now can be disabled at session level with use of a new variable query_response_time_session_stats.

Bugs Fixed

• Compilation warning was fixed for -DWITH_QUERY_RESPONSE_TIME=ON CMake compilation option, which makes QRT to be linked statically. Bug fixed #3841.
• A code clean-up was done to fix clang 6 specific compilation warnings and errors (bug fixed #3893, upstream #90111).
• Using -DWITHOUT_<PLUGIN>=ON CMake variable to exclude a plugin from the build didn’t work for some plugins, including a number of storage engines. Bug fixed #3901.
• A clean-up in Percona Server binlog-related code was made to avoid uninitialized memory comparison. Bug fixed #3925 (upstream #90238).
• Temporary file I/O was not instrumented for Performance Schema. Bug fixed  #3937  (upstream  #90264).
• A key_block_size value was set automatically by the Improved MEMORY Storage Engine, which resulted in warnings when changing the engine type to InnoDB, and constantly growing key_block_size during alter operations. Bugs fixed #3936, #3940, and #3943.
• Percona Server Debian packages description included reference to /etc/mysql/my.cnf file, which is not actually present in these packages. Bug fixed #2046.
• Fixes were introduced to remove GCC 8 compilation warnings for the Percona Server build, retaining compatibility with old compiler versions, including GCC 4.4. Bugs fixed #3950 and #4471.
• A typo in plugin.cmake file prevented to compile plugins statically into the server. Bug fixed #3871 (upstream #89766).
• -DWITH_NUMA=ON build option was silently ignored by CMake when NUMA development package was not installed, instead of exiting by error. Bug fixed #4487.
• Variables innodb_buffer_pool_populate and numa_interleave mapped to the upstream innodb_numa_interleave variable in 5.6.27-75.0 were reverted to their original implementation due to upstream variant being less effective in memory allocation. Now buffer pool is allocated with MAP_POPULATE, forcing NUMA interleaved allocation at the buffer pool initialization time. Bug fixed #3967.
• audit_log_include_accounts variable did not take effect if placed in my.cnf configuration file, while still working as intended if set dynamically. Bug fixed #3867.
• Synchronization between between innodb_kill_idle_transaction and kill_idle_transaction system variables was broken because of the regression in Percona Server 5.6.40-83.2. Bug fixed #3955.
• Executing the SHOW GLOBAL STATUS expression could cause “data drift” on global status variables in case of a query rollback: the variable, being by its nature a counter and allowing only an increase, could return to its previous value. Bug fixed #3951 (upstream #90351).
• ALTER TABLE … COMMENT = … statement caused TokuDB to rebuild the whole table, which is not needed, as only FRM metadata should be changed. The fix was provided as a contribution by Fungo Wang. Bugs fixed #4280 and #4292.
• A number of Percona Server 8.0 TokuDB fixes have been backported to Percona Server 5.6 in preparation for using MySQL 8.0. Bugs fixed  #4379,  #4380,  #4387,  #4378,  #4383,  #4384,  #4386,  #4382, #4391, #4390, #4392, and #4381.

TokuDB Changes and Fixes

• Two new variables, tokudb_enable_fast_update and tokudb_enable_fast_upsert, were introduced to facilitate the TokuDB fast updates feature, which involves queries optimization to avoid random reads during their execution. Bug fixed #4365.
• A data race was fixed in minicron utility of the PerconaFT, as a contribution by Rik Prohaska. Bug fixed #4281.
• Row count and cardinality decrease to zero took place after long-running REPLACE load, ending up with full table scans for any action.

Other Bugs Fixed

• #3818 “Orphaned file mysql-test/suite/innodb/r/percona_innodb_kill_idle_trx.result”
• #3926 “Potentially truncated bitmap file name in log_online_open_bitmap_file_read_only() (storage/innobase/log/log0online.cc)”
• #2204 “Test main.audit_log_default_db is unstable”
• #3767 “Fix compilation warnings/errors with clang”
• #3773 “Incorrect key file for table frequently for tokudb”
• #3794 “MTR test main.percona_show_temp_tables_stress does not wait for events to start”
• #3798 “MTR test innodb.percona_extended_innodb_status fails if InnoDB status contains unquoted special characters”
• #3887 “TokuDB does not compile with -DWITH_PERFSCHEMA_STORAGE_ENGINE=OFF”
• #4388 “5.7 code still has TOKU_INCLUDE_OPTION_STRUCTS which is a MariaDB specific construct”
• #4265 “TDB-114 (Change use of MySQL HASH to unordered_map) introduces memory leak”
• #4277 “memory leaks in TDB-2 and TDB-89 tests”
• #4276 “Data race on cache table attributes detected by the thread sanitizer”
• #4451 “Implement better compression algo testing”
• #4469 “variable use out of scope bug in get_last_key test detected by ASAN in clang 6”
• #4470 “the cachetable-simple-pin-nonblocking-cheap test occasionally fails due to a locking conflict with the cachetable evictor”
• #1131 “User_var_log_event::User_var_log_event(const char*, uint, const Format_description_log_event*): Assertion `(bytes_read == (data_written – ((old_pre_checksum_fd || (description_event->checksum_alg == BINLOG_CHECKSUM_ALG_OFF)) ? 0 : 4))) || ((“.

Find the release notes for Percona Server for MySQL 5.6.40-84.0 in our online documentation. Report bugs in the Jira bug tracker.

The post Percona Server for MySQL 5.6.40-84.0 Is Now Available appeared first on Percona Database Performance Blog.

Want to understand all the most important tech stats and trends? Legendary venture capitalist Mary Meeker has just released the 2018 version of her famous Internet Trends report. It covers everything from mobile to commerce to the competition between tech giants. Check out the full report below, and we’ll add some highlights soon. Then come back for our slide-by-slide analysis of the most important parts of the 294 page report.

• Internet adoption: As of 2018, half the world population, or about 3.6 billion people, will be on the internet. That’s thanks in large part to cheaper Android phones and Wifi becoming more available, though individual services will have a tougher time adding new users as the web hits saturation.
• Mobile usage: While smartphone shipments are flat and internet user growth is slowing, U.S. adults are spending more time online thanks to mobile, clocking 5.9 hours per day in 2017 versus 5.6 hours in 2016.
• Mobile ads: People are shifting their time to mobile faster than ad dollars are following, creating a $7 billion mobile ad opportunity, though platforms are increasingly responsible for providing safe content to host those ads.
• Crypto: Interest in cryptocurrency is exploding as Coinbase’s user count has nearly quadrupled since January 2017
• Voice: Voice technology is at an inflection point due to speech recognition hitting 95% accuracy and the sales explosion for Amazon Echo which went from over 10 million to over 30 million sold in total by the end of 2017.
• Daily usage – Revenue gains for services like Facebook are tightly coupled with daily user growth, showing how profitable it is to become a regular habit.
• Tech investment: We’re at an all-time high for public and private investment in technology, while the top six public R&D + capex spenders are all technology companies.

Mary Meeker, analyst with Morgan Stanley, speaks during the Web 2.0 Summit in San Francisco, California, U.S., on Tuesday, Nov. 16, 2010. This year’s conference, which runs through Nov. 17, is titled “Points of Control: The Battle for the Network Economy.” Photographer: Tony Avelar/Bloomberg via Getty Images

• Ecommerce vs Brick & Mortar: Ecommerce growth quickens as now 13% of all retail purchases happen online and parcel shipments are rising swiftly, signaling big opportunities for new shopping apps.
• Amazon: More people start product searches on Amazon than search engines now, but Jeff Bezos still relies on other surfaces like Facebook and YouTube to inspire people to want things.
• Subscription services: They’re seeing massive adoption, with Netflix up 25%, The New York Times up 43%, and Spotify up 48% year-over-year in 2017. A free tier accelerates conversion rates.
• Education: Employees seek retraining and education from YouTube and online courses to keep up with new job requirements and pay off skyrocketing student loan debt.
• Freelancing: Employees crave scheduling and work-from-home flexibility, and internet discovery of freelance work led it to grow 3X faster than total workforce growth. The on-demand workforce grew 23% in 2017 driven by Uber, Airbnb, Etsy, Upwork, and Doordash.
• Transportation: People are buying fewer cars, keeping them longer, and shifting transportation spend to rideshare, which saw rides double in 2017.
• Enterprise: Consumerization of the enterprise through better interfaces is spurring growth for companies like Dropbox and Slack.
• China: Alibaba is expanding beyond China with strong gross merchandise volume, though Amazon still rules in revenue.
• Privacy: China has a big opportunity as users there are much more willing to trade their personal data for product benefits than U.S. users, and China is claiming more spots on the top 20 internet company list while making big investments in AI.
• Immigration: It is critical to a strong economy, as 56% of top U.S. companies were founded by a first- or second-generation immigrant.

Salesforce just keeps on growing revenue. In another remarkable quarter, the company announced 3.01 billion in revenue for Q1 2019 with no signs of slowing down. That puts the CRM giant on a run rate of over $12 billion with the company’s most optimistic projections suggesting it could go even higher. It’s also the first time they have surpassed $3 billion in revenue for a quarter.

As you might expect Salesforce chairman and CEO Marc Benioff was over the moon about the results in the earnings call with analyst yesterday afternoon. “Revenue for the quarter rose to more than $3 billion, up 25%, putting us on $12 billion revenue run rate that was just amazing. And we now have $20.4 billion of future revenues under contract, which is the remaining transaction price, that’s up 36% from a year ago. Based on these strong results, we’re raising our full year top line revenue guidance to $13.125 billion at the high end of our range, 25% growth for this year,” Benioff told analysts.

Brent Leary, an analyst who has been watching the CRM industry for many years, says CRM in general is a hot area and Salesforce has been able to take advantage. “With CRM becoming the biggest and fastest growing business software category last year according to Gartner, it’s easy to see with these number that Salesforce is leading the way forward. And they are in position to keep moving themselves and the category forward for years to come as their acquisitions should continue to pay off for them,” Leary told TechCrunch.

Bringing Mulesoft into the fold

Further Benioff rightly boasted that the company would be the fastest software company ever to $13 billion and it continued on the road towards its previously stated $20 billion goal. The $6.5 billion acquisition of Mulesoft earlier this year should help fuel that growth. “And this month, we closed our acquisition of MuleSoft, giving us the industry’s leading integration platform as well. Well, integration has never been more strategic,” Benioff stated.

Salesforce CEO Marc Benioff Photo: TechCrunch

Bret Taylor, the company’s president and chief product officer, says the integration really ties in nicely with another of the company’s strategic initiatives, artificial intelligence, which they call Einstein. “[Customers] know that their AI is only as powerful as data it has access to. And so when you think of MuleSoft, think unlocking data. The data is trapped in all these isolated systems on-premises, private cloud, public cloud, and MuleSoft, they can unlock this data and make it available to Einstein and make a smarter customer facing system,” Taylor explained.

Leary thinks there’s one other reason the company has done so well, one that’s hard to quantify in pure dollars, but perhaps an approach other companies should be paying attention to. “One of the more undercovered aspects of what Salesforce is doing is how their social responsibility and corporate culture is attracting a lot of positive attention,” he said. “That may be hard to boil down into revenue and profit numbers, but it has to be part of the reason why Salesforce continues to grow at the pace they have,” he added.

Keep on rolling

All of this has been adding up to incredible numbers. It’s easy to take revenue like this for granted because the company has been on such a sustained growth rate for such a long period of time, but just becoming a billion dollar company has been a challenge for most Software as a Service providers up until now. A $13 billion run rate is in an entirely different stratosphere and it could be lifting the entire category says Jason Lemkin, founder at SaasStr, a firm that invests in SaaS startups.

“SaaS companies crossing $1B in ARR will soon become commonplace, as shocking as that might have sounded in say 2011. Atlassian, Box, Hubspot, and Zendesk are all well on their way there. The best SaaS companies are growing faster after $100m ARR, which is propelling them there,” Lemkin explained.

Salesforce is leading the way. Perhaps that’s because it has the same first-to-market advantage that Amazon has had in the cloud infrastructure market. It has gained such substantial momentum by being early, starting way back in 1999 before Software as a Service was seen as a viable business. In fact, Benioff told a story earlier this year that when he first started, he did the rounds of the venture capital firms in Silicon Valley and every single one turned him down.

You can bet that those companies have some deep regrets now, as the company’s revenue and stock price continues to soar.  As of publication this morning, the stock was sitting at $130.90, up over 3 percent. All this company does is consistently make money, and that’s pretty much all you can ask from any organization. As Leary aptly put it, “Yea, they’re really killing it.”