We at FCS are good at what we do. Very good. That’s why people call us to ask for help after problem backups, how do they make it better?
One thing that stands out when we do the analysis of these calls is whatever means the prospect has used to backup it isn’t this bit that failed, it’s only at the recovery stage that they find a failure.
That’s regardless of the testing people do, verifications, restores etc.
When something goes wrong and it’s no longer an exercise to see if things look OK, reputations of strategy, product and people are on the line. Did the right decision get made 2 years ago, was the cost/risk analysis accurate? What went wrong? And the most fundamental of all questions, can I get my data back?
Unfortunately, too often the answer is ‘some of it’ and that’s from a combination of tapes, local copies made by people, even old laptops and USB sticks used to transport the information in the first place have come up trumps for some customers.
But shouldn’t we be turning all that on its head and rather than looking at the backup costs and method, look at the restore phase? After all, if being able to restore the data is the target then the backup has to be right from the outset.
Ask your provider, ‘Can you 100% guarantee this platform will allow me to recover from any disaster?’
If they answer ‘yes’ then they don’t know what they are doing, if they offer a test regime and a strategy with support and evidence that it has always worked so far, then you’re nearly there. The only proof will be their response during a disaster and how well the proposed solution copes with your data volumes, rates of change and giving you your data back when you need it.
Let’s start with a look at the issues…
Data is growing, and fast. We’re being told that we are generating more data exponentially, for instance, by 2020 we will be creating 1.7Mb per second for every inhabitant of the planet!
So, we have data and are generating more every second. We all sit at our computers writing blog posts, reports and emails all the time, creating data. Predictions are that we will send 246 BILLION emails per day by 2019.
Typically, the more we depend on IT the more it costs when it fails, regardless of the reason.
One key take away is that we’ve generated more data in the past two years than the entire history of the human race! Ok, not all of it equates to Cicero or Shakespeare - and I would imagine a great deal of it is pictures of fluffy kittens in a large wine glass - but it is being generated at a massive rate.
Not everyone runs a business the size of Amazon.com but the data you have is estimated to be growing at 39% each year and it’s there for a reason, to support your business. Actually, it might even shrink a little with GDPR (General Data Protection Regulation, the EU wide regulation designed to let the law catch up with fast developing IT systems, data collection and sharing) enforcement looming but your data needs to be available to the people and applications that need it. There will be more on GDPR in our next post.
Isn’t making data available simply just putting a bunch of disks on the network that’ll be fine?
Not really. While your data is, on average, growing at 39% per year it is growing 9% faster than we are creating data, pointing to massive duplication. Unstructured data is open to abuse and there is a storage management problem caused in the main by users being scared of deleting anything. When I am delivering a presentation, I will have emailed a copy to the event organiser, have a copy on my laptop, one on a USB stick and possibly another on my phone and one in the cloud, just to be certain. Over the top, possibly, but a strategy based on many years of presenting and suffering power, luggage, venue and organiser failures, not to mention my own memory!
The growth and security of the data you have adds complex challenges across a business. GDPR (there is GDPR again, it will be covered in a future blog post – and it’s a big topic as I hope you realise by now) demands that data is kept only for as long as it is required and any data kept must be kept in accordance with GDPR. Currently, all member countries have their own laws and regulations and GDPR is intended to harmonise those into a single version. So, you need to understand what data you hold, here it resides and respond to ‘right to be forgotten’ requests quickly. Across your entire organisation. Including SAN or NAS, servers, tape archives, everything.
But I digress, and there are more traditional issues faced everyday by business…
Disaster.
This is a broad category including flood, fire, accident (I do remember being called out as an engineer in the middle of the night to a customer with a stolen car embedded in the wall to the server room, the rack bent in half and servers, switches etc were all unsurprisingly ruined). The weather is, for whatever reason, getting worse and floods more frequent. Leeds city centre, among others, was hit hard in late 2015 and power was lost across a wide area – even those buildings not affected by flood. Fires are more frequent than we’d like and even if it doesn’t affect the equipment directly, water damage to the building either from the sprinkler system or firemen, can cause internal floods and again, ruin equipment.
Malware
This is very difficult to protect against because no one knows what the next threat will be. Right now Ransomware is growing at an alarming rate and it is proving difficult to protect against. We recently published a news item giving our view on Ransomware, the article can be found here, of course we are fairly black and white, we like to be able to offer a solution that will 100% guarantee to defend against Ransomware and other nasties, take a look it’s quite interesting.
Equipment Failure
That sinking feeling when you’re hoping to work on a presentation over the weekend so you can celebrate your other half’s birthday and can’t access the VPN, or that panic call from the boss as you are half way to the office on Monday morning to tell you nothing works. If you’re lucky it’s the mains that tripped or the UPS or a switch failed. If your unlucky it’s a firmware update gone wrong or server failure that a reboot will fix. If you’re really unlucky it could be a lot worse.
Theft.
Though a fortunately rare event these days, this does still happen. Data centres are subject to sophisticated attacks, more frequently our offices broken into and equipment stolen. The children’s charity Plan UK had equipment stolen from their own offices giving rise to data protection concerns.
Almost all data centres have incredibly tight security (though it hasn’t always been like that!) and in recent years it’s often harder to get in to one than it is Fort Knox so it could be argued that the risk from theft would be greatly reduced by either collocating your servers or full cloud adoption in such a secure data centre.
Human Error
This is the most common cause and one that prompts most of the Friday funnies that circulate, allowing those of us who didn’t do anything like it ever (honest!) to sit back and thank the great bit bucket in the sky that it wasn’t, for once, us… accidentally deleting masses of data is remarkably easy and the speed our systems work at now gives us less time to realise our mistakes. This one has an almost successful conclusion, however, there is an important lesson here.
The first article stated “out of 5 backup/replication techniques deployed none are working reliably or set up in the first place.” Really? None of the backup systems were configured correctly or worked reliably? They were tested, weren’t they? And…? If you aren’t 100% sure your backups are working right you need to find out why. And how to do it better.
The future’s so bright scarey I gotta wear shades armour and asbestos underwear…
So, we have a nice little ‘storm of things’ brewing in the next few years that seem to add up to an insurmountable problem. Infrastructure fails for whatever reason, regulation around information governance is forcing changes to people’s IT management and data gathering and retention strategies. Surely, this is a difficult problem to solve… isn’t it?
Err… Well, not necessarily. In fact, it’s quite easy.
Let’s get back to some basics. FCS take Disaster Recovery and Business Continuity planning very seriously and we don’t treat them as isolated practices. Rather they go together like a hand in a glove and if one is right then it follows the other is right too.
Additionally, if we take one of the first points made in this blog, that focussing on backup is wrong and focussing on the restore is the path to nirvana, the moving this backup requirement to an expert MSP is a clever move and can add lots of features to your IT provision for not much money.
When you use FCS as an MSP we can look after your data backups for a simple monthly fee.
We setup and support the solution so it is done correctly, and tested, by our experts and we monitor the solution 24x7. We even support you 24x7 as part of the cost. No more tape, just a simple service that lets you call up and ask for a file, set of files or everything you ever had. Additionally, you pay for what you use, stored de duplicated, compressed data. No blocks of storage you are half using.
We know it will be available because every time we pull the changes to your data in, we test them. We do scheduled test restores of the data on a regular basis and every day you get an email telling you how restorable your data is. If anything is amiss we work with you to help rectify the situation (server rebooted, doing an update, unplugged by the cleaner, disk error – all the usual causes) so you get an early warning something went awry.
We support a great many operating systems (Windows, Linux, Unix, MAC OS/X, OS/400 etc.), varied environments (AWS, Azure) and can backup common hypervisors such as Microsofts Hyper-V, VMware and Xenserver. Specific point server products, like Exchange, Sharepoint, MySQL, MS SQL, PostgreSQL and Oracle are all supported with no additional agents required in most use cases. If you mix cloud applications, Office 365, Google Apps (or G Suite) and SalesForce, we can back that up too.
Cloud storage also offers long-term retention options that local storage can’t always manage. Whereas local storage is a relatively finite resource (with less capacity available), cloud storage is the opposite with virtually unlimited capacity. This opens the door to long-term retention policies where necessary, which can be tailored to meet your requirements.
Outsourcing your requirements can lower costs and bring a skill set into your data protection plan that would be both difficult and expensive to resource. Additionally, how much time does your IT team spend backing up, test restoring, checking and fixing errors, replacing tapes or external drives? Wouldn’t it be better to simply trust an expert? We don’t learn medicine when we break our leg, we go to a doctor and put our trust in them. We’d say your data was at least valuable enough to trust to an expert.
Security is built in from the start as we use AES-256 Encryption and our solution has NIST FIPS 140-2 certification. Like any other asset your company owns you want to make sure the data is only accessible by you, that’s why you choose your own encryption key and you don’t disclose it to anyone outside your company. The data centres we use are Tier III. We are happy that’s secure enough for our hardware and platforms and your data. We do security and do it well.
Every business has a cost of downtime (ask the CFO what would happen if the lights went out for a three hour period and watch his face pale), and recovery objectives are made based on this. We use a platform that can meet your objectives, and we tailor the service to meet your requirements.
Short Recovery Time Objectives (RTOs) are an important part of the modern IT environment. To ensure we can meet these, we use Disaster Recovery products that are renowned in the industry for offering outstanding results.
Just as important as the recovery time is the Recovery Point Objective (RPO). When recovering infrastructure or data, getting the right version back is essential to bringing your environment back to life. Our DR solution offers granular restores both at the infrastructure, and the application level so that you can recover exactly what you need.
If you want to replicate your environment in our DC and spin it up in and emergency then we can, and do, provide that service. We have experts on hand to guide you through the process of a (hopefully) test invocation or if the worst happens, the real thing. We know what to expect, what is normal and can keep your blood pressure under control, we understand the process and provide reassurance, guidance, and ways to improve your own process and documentation to make a real disaster much easier to cope with.
Today FCS supports customers in many sectors with our fast and flexible, cost effective solution to organisations across the UK.
• ISO 27001 accredited company
• UK only, mirrored Tier 3 BS5979 & BS8418 Accredited Data Centres
• One size fits no-one, everything is designed to your requirements reducing costs as a bonus
• Incremental forever snapshots
• Focus on restore, not backup
• Daily report assists with compliance
• 24x7x365 UK based support
• Support for Windows, VMware, HyperV, Solaris, AIX (most versions), OS/400 and more
• Free initial consultancy, POC testing and installs
• De-duplicated, Compressed and Encrypted – you keep the keys so only you can see the data
• The only way to recover from Ransomware quickly
We will keep it as safe as we would our own, in fact we have our own data there too so we have to! And here is an important point. We have never lost any customer data. Not a single bit. Ever. Oh, and a single call can go some way to satisfying your GDPR needs too. Being able to delete from the store makes tape all but obsolete. Or at least a very, very expensive alternative.
So, it’s time to stop worrying about backups and focus on restores, that’s where the pain will be in coming years. Prepare now by moving to a managed cloud service and let FCS take the stress out of IT.
Our next blog is a quick look at GDPR and it’s going to be another interesting one.
FCS. Make us your first resource, not your last resort…