With the general availability of Google Cloud Platform’s latest database offerings — the second generation of Cloud SQL, Cloud Bigtable, and Cloud Datastore — Google is setting up a cloud database strategy founded on a basic truth of software: Don’t get in the customer’s way.
For an example, look no further than the new iteration of Cloud SQL, a hosted version of MySQL for Google Cloud Platform. MySQL is broadly used by cloud applications, and Google is trying to keep it fuss-free — no small feat for any piece of software, let alone a database notorious in its needs for tweaks to work well.
Most of the automation around MySQL in Cloud SQL involves items that should be automated anyway, such as updates, automatic scaling to meet demand, autofailover between zones, and backup/roll-back functionality. This automation all comes via a recent version of MySQL, 5.7, not via an earlier version that’s been heavily customized by Google to support these features.
The other new offerings, Cloud Datastore and Cloud Bigtable, are fully managed incarnations of NoSQL and HBase/Hadoop systems. These systems have fewer users than MySQL, but are likely used to store gobs more data than with MySQL. One of MySQL 5.7’s new features, support for JSON data, provides NoSQL-like functionality for existing MySQL users. But users who are truly serious about NoSQL are likely to do that work on a platform designed to support it from the ground up.
The most obvious competition for Cloud SQL is Amazon’s Aurora service. When reviewed by InfoWorld’s Martin Heller in October 2015, it supported a recent version of MySQL (5.6) and had many of the same self-healing and self-maintaining features as Cloud SQL. Where Google has a potential edge is in the overall simplicity of its platform — a source of pride in other areas, such as a far less sprawling and complex selection of virtual machine types.
Another competitor is Snowflake, the cloud data warehousing solution designed to require little user configuration or maintenance. Snowflake’s main drawback is that it’s a custom-build database, even if it is designed to be highly compatible with SQL conventions. Cloud SQL, by contrast, is simply MySQL, a familiar product with well-understood behaviors.
setup procedure will allow users to configure SQL Server on Microsoft Azure without the aid of a database administrator.
“The new wizard for building and configuring a new virtual machine with SQL Server 2014 is very well put together,” said Denny Cherry, founder and principal consultant for Denny Cherry and Associates Consulting. “It helps solve a lot of the complexity of building a new SQL Server, specifically around how you need to configure the storage in order to get a high-performing SQL Server VM.”
Joseph D’Antoni, principal consultant at Denny Cherry and Associates Consulting, said that one of the major challenges with Azure was allocating storage. For instance, he said, to configure SQL Server on an Azure VM, you needed to allocate disks manually to get the needed amount of IOPS. This meant you had to know exactly what your storage application needs were for optimal performance, and many people were “kind of guessing,” D’Antoni said. With the new wizard, all you have to do is enter the required number of IOPS and storage is allocated automatically.
Automating SQL Server setup for an Azure VM means that no longer does everything have to be done manually: Connectivity, performance, security and storage are configured automatically during setup. “I think it does simplify what was a pretty complex process,” D’Antoni said.
You can now use the Internet to set up SQL Server connectivity and enable SQL Server authentication through the Azure Web portal. Previously, connecting SQL Server to an Azure VM via the Internet required a multistep process using SQL Server Management Studio. The new automated configuration process lets you pick whether to expand connectivity to the whole Azure virtual network or to connect only within the individual VM.
The new process for configuring SQL Server on an Azure virtual machine also includes automated patching and automated backup. The automated patching allows you pick a time when you want all your patches to occur. Users can schedule patches to minimize the impact they’ll have on the workload. Automated backup allows you to specify how long to keep backups.
“I think that these are a great enhancement on the old process of having to know how to configure these components manually within the VM,” Cherry said, “because these configurations can get tricky to configure.”
D’Antoni added that this innovation is going to affect smaller organizations the most, because it means that they won’t need an expert to move SQL Server onto an Azure virtual machine. “[The simplified configuration] gives the power to someone who is deploying a VM when they would have needed an administrator or a DBA before. To that extent, it’s kind of a big deal.”
SQL Server 2016 is about to launch with a long list of shiny new built-in features along with much-needed improvements…
to important but humdrum capabilities database administrators rely on.
The upcoming release, slated for June 1, marks Microsoft’s initial go at a cloud-first version of SQL Server. It also happens to be one of the biggest releases in its history, with something for everyone, said Andrew Snodgrass, aresearch vice president at Directions on Microsoft, an independent analysis firm in Kirkland, Wash.
Some of the most notable SQL Server 2016 features include performance tuning, real-time operational analytics, visualization on mobile devices, and new hybrid support that allows admins to run their databases on premises and on public cloud services. Microsoft also invested in less sexy, but important, SQL Server 2016 features that hadn’t been improved in some time.
SSRS and SSIS finally get some love
Indeed, SQL Server 2016 is an exciting release for reporting and ETL practitioners, according to Tim Mitchell, principal atTyleris Data Solutions, a data management services provider in Dallas.
SQL Server Reporting Services (SSRS), long suffering from release after release of few remarkable changes, received a significant makeover, he said. The classic Report Manager interface has given way to a brand new portal that looks and acts like a modern Web app — and it’s brandable, he noted.
The new KPI functionality makes building dashboards much easier, and the mobile reporting tools Microsoft added from the 2015 Datazen acquisition have made SSRS relevant for companies that support reporting for mobile users, according to Mitchell.
Andrew Snodgrassresearch vice president, Directions on Microsoft
The changes in SQL Server Integration Services (SSIS) are more subtle, but significant. When the SSIS catalog was introduced in 2012, it brought many changes but one significant limitation: SSIS packages could no longer be deployed individually; instead, the entire project had to be deployed at once, said Mitchell, who is also a data platform MVP.
“To their credit, Microsoft heard the roar of negative feedback and have changed this in 2016, once again allowing package-by-package deployment,” he said.
For those boxed in by the limitations of SSIS catalog logging, a new feature that supports custom logging levels brings freedom. Also, for those who were previously forced to install multiple versions of SQL Server Data Tools to support the various versions of SSIS, the new SQL Server Data Tool designer allows for targeting of a specific SQL Server Integration Services version when developing SSIS projects, Mitchell said.
Performance tuning, In-Memory OLTP and PolyBase
Perhaps the most useful SQL Server 2016 feature for database administrators involves performance tuning, which allows DBAs to monitor and record the full history of query execution plans to diagnose issues and optimize plans. It will be invaluable for upgrades and patching to see where changes have impacted performance, Directions on Microsoft’s Snodgrass said.
“Performance tuning with the new Query Store is one of those ‘about time’ solutions,” he added.
Other notable improvements to SQL Server 2016 are PolyBase integration, and performance features with In-Memory OLTP and columnstore indexes are finally mature enough for most companies to deploy them, according to Snodgrass.
“The supported feature set, as compared to on-disk tables, was not on parity and it made it difficult to migrate to In-Memory tables without a great deal of effort,” he said.
In addition, Microsoft raised the size limit on memory-optimized tables to 2 TB, and those memory-optimized tables can be edited. Another important SQL Server 2016 feature is the ability to combine In-Memory OLTP and columnstore indexes on a single table.
“It’s not for everyone, but there are cases where it would be great to have real-time statistics and trends available from live, transactional data,” Snodgrass said. “Right now the process is time-delayed, since it usually requires grabbing transactions at a point in time and performing analysis somewhere other than on the transactional table.”
However, Snodgrass cautioned, DBAs shouldn’t try this without the proper infrastructure. “You’d better have beefy equipment and failover configured before trying this,” he said.
PolyBase, which provides the ability to access unstructured data in Hadoop, has been in specialized versions of SQL Server since 2012. It will be included in SQL Server 2016 Enterprise edition. That means organizations that didn’t want to spend the money on big equipment can now use existing SQL Server installations to pull unstructured data, Snodgrass said.
“Of course, that doesn’t immediately solve the problem of deploying Hadoop, but it is good for the SQL guys,” he added.
JSON, live queries and analytics
JSON support is an important feature because it allows users to read and write JSON-based documents. This provides a controlled gateway for sharing organizational data with more mobile platforms. Companies have struggled to write database apps for mobile devices, because the data storage options weren’t compatible with on-premises data platforms, Snodgrass said.
“This provides a much easier method for transporting that data between mobile/Web solutions and relational database applications,” he said.
Other SQL Server 2016 features users are excited about are Query Store, Live Query Statistics and Live Query Plans (in Management Studio), according to Gareth Swanepoel, a senior data platform and BI consultant at Pragmatic Works Inc., a SQL Server software and training provider in Middleburg, Fla.
“These [Query features] represent a major improvement to performance tuning on a system,” Swanepoel said. “DBAs will have access to vastly enhanced metrics.”
In addition, SQL Server Management Studio’s release schedule has been separated from the main SQL Server releases, and it will be updated more frequently than before.
Perhaps least impressive of the new SQL Server 2016 features, according to Snodgrass, is SQL Server R Services, which supports advanced analytics with the R programming language.
“The ability to incorporate R scripts in stored procedures is interesting, but the audience is very limited and other tools out there do a good job of this,” he said. “It’s important for the long term, but I suspect adoption will be slow in the beginning.”
SQL Server 2016 editions will include Enterprise, Standard, Express and Developer. The SQL Server 2016 Developer edition, with the full capabilities of the latest SQL Server release, will be free.
A new plan proposed for OpenJDK would give Java a local-variable type inference capability so that it’s easier to write code while continuing with static type safety.
The measure was proposed earlier this month in JEP (JDK Enhancement Proposal (JEP) 286. Documentation currently does not cite a version of Java where the improvement would be featured.
“We seek to improve the developer experience by reducing the ceremony associated with writing Java code while maintaining Java’s commitment to static type safety by allowing developers to elide the often-unnecessary manifest declaration of local variable types,” the proposal states. Other statically typed curly braced languages, such as C++, Scala, and Go, already support some variant of local-type inference. “Java is nearly the only popular statically typed language that has not embraced local-variable type inference,” the JEP stresses. “At this point, this should no longer be a controversial feature.”
Java developers frequently complain about the volume of boilerplate coding required, according to the JEP. “Manifest type declarations for locals are often perceived to be unnecessary or even in the way; given good variable naming, it is often perfectly clear what’s going on,” the JEP states. “The need to provide a manifest type for every variable also accidentally encourages developers toward overly complex expressions; with a lower-ceremony declaration syntax, there is less disincentive to break complex chained or nested expressions into simpler ones.”
The plan would be restricted to local variables with initializers, indexes in the enhanced for-loop, and locals declared in a traditional for-loop. “Quantitatively, we want that a substantial percentage of local variable declarations in real code bases can be converted using this feature, inferring an appropriate type.”
OpenJDK serves as an open source Java SE implementation and is backed by Oracle.Java SE 9, Oracle’s official version of the next generation of the platform, is due in a year. The JEP has the endorsement of Oracle’s Mark Reinhold, chief architect of the company’s Java platform group.