On 29 October 1969, a team of American university researchers sent a message containing the letters “LO” across a network spanning just a few miles. They had intended to spell out the word “login”, but the system crashed with the entry of the letter “G”.
Despite this haphazard start, the tiny network evolved into what we now call the internet. Nearly fifty years on, the web supports 500 million tweets, 5 billion searches and nearly 300 billion emails a day. But it’s not just the internet giants which have to handle such vast sums of data.
As consumers’ appetite for digital services has grown, companies have been forced to grapple with a proliferation of databases, as well as the platforms, structures, infrastructures, editions and IP models which support them.
This issue is exacerbated by the adoption of new development projects, skunk-works efforts and mergers and acquisitions. Organisations are increasingly turning to corporate IT to own the databases and to database administrators to manage them. So what are the key challenges in managing cross-platform databases, and what steps can be taken to overcome them?
1. Consider using monitoring software
A recent survey by Unisphere Research revealed that 72 per cent of DBAs are managing an increasing number of databases. Meanwhile, nearly 70 per cent of administrators have to contend with more than one database platform and nearly a third have to manage three or more.
Claims that “we’re an Oracle shop,” or “we run only SQL Server” are decreasing in number because Oracle, for instance, has no SQL offerings for big data while Microsoft has on-premises and cloud offerings alongside its offerings for big data.
This isn’t an issue for everyone. A sufficiently large company may be able to afford single-platform DBAs who can focus 90-plus per cent of their time on one product or another. But in most cases, companies are asking their DBAs to manage multiple databases.
Doing so may be harder than it seems. Most DBAs consider themselves experts in one primary platform. When cutting their teeth on a new, secondary platform, they tend to think: “How do I do this on MongoDB and make it similar to the way I’m accustomed to doing it on SQL Server?”
Before long, they trip up on the differences between the platforms and the gaps in their own knowledge, training or experience. If you’re a DBA managing three database platforms, you may find yourself juggling three different sets of monitoring tools, each of which presents information in a different way with different a user interface (UI).
Foglight for Databases is the database monitoring solution from Quest Software that is designed for cross-platform environments. Foglight monitors traditional database platforms like Oracle, SQL Server and DB2, plus newer platforms, both open source and non-relational, like Sybase/ SAP Adaptive Server Enterprise (ASE), MySQL (including MariaDB, Percona and AWS Aurora), PostgreSQL (including EnterpriseDB), MongoDB (all editions) and Cassandra (open source Apache and commercial DataStax).
Its cross-platform coverage centralises monitoring under a single umbrella. Foglight simplifies the UI of examining the health of databases, regardless of platform, by enabling drill-down from one layer to the next through clicks. DBAs enjoy a uniform view of statistics on instance performance, database availability, disaster recovery preparedness, remaining storage and server resources (CPU, memory, disk latency). The tasks of database tuning and workload analysis become consistent across database platforms, making life easier for DBAs.
2. Avoid paying extra for diagnostics packs
Some vendors offer several editions of their databases, such as a fully featured enterprise edition, a lower-cost standard edition with fewer features or limited functions, and a stripped-down express edition. But even if you’ve paid for a top-tier package, some diagnostics packs will require an additional licensing fee.
For DBAs in cost-conscious organisations, that can be frustrating. When they have limited diagnostic capability, they have fewer tools for resolving problems and less control over their databases. The Performance Investigator (PI) feature included in Foglight is designed to address that frustration. PI does not access any tables, views, procedures or APIs that are subject to licence, so DBAs can use the product to get all the performance diagnostics they need without licensing more-expensive editions of the databases.
3. Automate time-consuming processes
No matter how diligently DBAs try to consolidate or decommission obsolete servers, new servers still seem to come on line faster. Almost 40 percent of DBAs manage more than 25 database instances each, and most DBAs believe that the number of databases in their purview is rising, according to Unisphere’s research.
Besides managing inventory, admins have to manage capacity of their IT infrastructure and keep an eye on areas of over- or under-utilisation. Striking the right balance is difficult enough in small environments, let alone in large ones. This is where automation can play a part. Patching, maintenance, backup testing and performance management tasks can all be automated, freeing up DBAs for higher value work.
Foglight automates a rich set of out-of-the-box dashboards and reports on database environments of any size. This means DBAs can sit separately, examining the same Foglight dashboard simultaneously for underutilised systems ripe for consolidation. Or, in the case of automated reporting for upper management, DBAs can schedule regular reports on, say, recent service-level agreement (SLA) attainment.
4. Create a single view for all your databases
Gone are the days when all databases resided on disks spinning in racks on the company’s premises. Variety in the location and type of servers has become a fact of life for DBAs charged with monitoring databases.
Besides being located on premises or in the cloud, databases can reside on physical or virtual servers. In global organisations they can be anywhere on the planet, and DBAs must manage them remotely. Managed service providers (MSPs) administer databases on behalf of their customers. In all such cases, DBAs are tasked with efficiently monitoring and diagnosing the performance of far-flung database environments.
Foglight is designed to provide a single view with the same look and feel for all databases regardless of location and physical/virtual status. It monitors databases remotely without the need for additional schema objects in the database or an installed agent on the host computer or database server. Its agentless architecture allows it to monitor any database environment from anywhere — from on-premises to the cloud, from cloud to cloud and from cloud to on-premise — in all architectures and configurations.
5. Don’t overburden the system
The cardinal rule in database monitoring is to not do more harm than good. Given that most performance monitoring techniques introduce some performance hit of their own, DBAs are keen to monitor as thoroughly as possible with minimal overhead. The larger the environment, the greater the overall effect of overhead-heavy monitoring.
Foglight is built around carefully chosen APIs for capturing performance data. To keep its overhead as low as possible, it does not use computationally expensive APIs. It also stores all of its monitoring and diagnostics data in centralized repositories. Nothing that Foglight collects goes back onto the database server itself; DBAs running diagnostics or performance tuning are working exclusively against Foglight.
As database technology blossoms from on-premises environments to the cloud and DBaaS, DBAs face an increasingly fragmented landscape of database monitoring. It is nearly impossible to be an expert in traditional, NoSQL, relational, non-relational and open source platforms simultaneously, yet the need to have a constant, homogenous handle on a changing, heterogeneous environment continues to grow, especially as organizations run more and more databases.
Foglight for Databases gives DBAs deep insight into the health and performance of their databases regardless of platform. It abstracts the difficult work of choosing and calling the APIs best suited to reporting on each type of database and rolls results into a single UI so DBAs can monitor and tune their environment efficiently.
Jason Hall is currently a senior sales engineer at Quest focused on database solutions.
Pini Dibask is a Quest product manager for database performance management solutions.