Four signs your DBA might need a hand

Jon Cowling 06-Jul-2018 09:39:20

Four signs your DBA might need a hand

Have you ever met a DBA who didn’t look overworked? Keeping the lights on and everything running smoothly can be a very stressful job, and we’ve highlighted four key areas where DBAs often call us in for a hand in this article. If any of these resonate with your organisation and you’d like to talk it through further, why not book a free consultation with us? There’s a link at the bottom of this page.

You’re thinking of moving to the cloud

Everybody seems to be going cloud-first these days, right? They are, but it doesn’t mean it’s an easy task, or that the traditional concerns over performance, availability and security aren’t valid. There are lots of options to consider when trying to design the cloud technology stack, and an outright “lift and shift” replication from an existing on-premise system to the cloud isn’t always the best idea, because any problems in the existing on-premise configuration will be replicated, and possibly exacerbated. If it’s down to us, we like to be involved in cloud migration discussions early-on, so we can assess your environment and get a good understanding of the requirements and recommend the best course of action to proceed.

Given cloud databases haven’t been around all that long, we’re helping clients understand the security implications of cloud-hosted environments, as well as data platform modernisation to optimise their systems for peak performance. It’s entirely possible too that a fully cloud solution might not be best - like in the manufacturing industry, for instance, where the latency involved would be too high, so parts of your infrastructure might be on premise. Often, the cloud databases aren’t exactly the same as the on-premise versions your IT teams may be familiar with - last week we were asked for help with scheduling SQL Server native jobs, which you can’t actually do using the PaaS solution the customer had provisioned in their cloud of choice, so we had to figure out a couple of workarounds that would fulfil the client’s requirements.

You’ve had a database outage

This is the dreaded scenario, and for most people it doesn’t happen all too often. Database outages are most commonly caused by hardware failure, which could be anything from a single hard drive failing up to a whole server - or any of its component parts. Storage issues can cause database corruption, which leads to crashes and outages, and user error can account for a lot of outages too. The configurations can get pretty complicated to setup and if you’re not familiar with it, it’s easy to overlook something crucial. We were recently asked to help tackle an outage for a client in the banking industry who’d lost their backup drive, or so they thought. Unfortunately the internal IT team had accidently placed live production database files onto the backup drive, when the drive failed, the database wouldn’t respond, and it was an internal system that processed loan applications, so pretty crucial. They gave us a call and we managed to switch them over to a DR copy we’d set up for them and they were back up and running smoothly within an hour.

Thankfully, we’ve got amongst the best DBAs in the country, and we can get into your system, understand it and help you restore your environment as fast as possible. If you’d like us to, we can analyse the causes of the issue and take action to reduce the chances of anything like that happening again, too.

Your staff are reporting poor application performance

Poor application performance is one of the most common signs that your current database configuration is not coping well. You’re in great hands with DSP-Explorer, though - we get asked to look at optimising application performance every single day, and often it’s just a question of taking a fresh look at the setup of the database environment and making sure the common parameters are set properly, and that disk load is being spread effectively. Other times the application might be querying the database in a really inefficient manner, and there might be ways we can redesign requests to streamline performance.

This can have serious real-world implications beyond increased loading time. Our team recently helped a customer in the manufacturing industry with a large production line and associated warehouse operations. They called us to say that their Oracle database had gone from processing each item in five seconds to four minutes and had been experiencing this slowdown for three days by the time they called us. Given the trend for lean manufacturing, any production slowdown has implications far down a supply chain, and this customer weren’t able to meet their orders.  We were able to respond in four hours, drafting up a discovery session to examine their databases, isolating the issue and fixing the problem within eight hours. In the process, we managed to cut their processing time down to four seconds, meaning their production lines would no longer bottleneck as a result of the database being overloaded.

Your DBA wants to go on holiday!

Adding extra capacity to support your DBA does more than help top up their tan and maintain their sanity, though. We all know how hard it is to recruit (and keep!) experienced DBAs, and if your organisation is already stretched, we can help with out-of-hours support, or ad-hoc requirements that crop up. We handle lots of short-term projects for clients, usually around DR and backup configurations, helping figure out Oracle DataGuard or SQL Always-On, and clustering with RAC and other technologies. With the GDPR regulations having just come into force, lots of customers are looking for help configuring TDE (Transparent Data Encryption), too. We had one organisation last month who’d enabled TDE across all 40 of their servers, and later realised that it meant their backups were encrypted and thus couldn’t refresh their test environment from a backup. It’s things like that where it helps to know the knock-on effects, so you can mitigate them rather than have a nasty surprise down the line.

Want to get us to take a look?

Whether your database is cumbersome, causing frequent outages, letting overnight batch jobs spill into the working day or just eating up your IT budget in constant support, we’re ready to come to your rescue.

We encounter challenges like these every day. They’re issues we absolutely know how to solve, and we love them, even if you don’t. So, don’t let your database limp on – let us at it. We’ll make good your database woes and while we’re there, make a truly positive impact on your business.

Book a consultation with us here and we’ll see how we can help you today.

New call-to-action