Any solution provider struggling to have their customers understand the importance of disaster recovery need only remind them of last year’s Hurricane Sandy in the US.
The Category 3 hurricane generated terrifying winds and a resulting surge in sea levels in the south as blizzards beat the north.
Power was cut from Virginia to New England, crippling one of the most densely populated areas of the country as supply chains were thrown into disarray.
Dubbed ‘Frakenstorm’ by the US media, Sandy flooded businesses in downtown Manhattan and forced the New York Stock Exchange to close for the first time in 30 years.
United Airlines lost $US90 million in flight cancellations. Pharmacy giant Walgreens had to close 1,400 stores. US telcos reported to the FCC a quarter of all mobile phone sites across 10 states went down while cable operators reported 25 percent broadband outages.
All up Sandy cost the US economy $US70 billion.
Closer to home, the earthquakes that rocked the New Zealand city of Christchurch in 2010 and 2011 had a devastating impact, even for a country well accustomed to the vagaries of tectonic plates. Important businesses including one of the country’s biggest TV stations were reduced to rubble, while significant parts of the city remain condemned and off limits.
Engineers and staff at NZ services provider CCL were as busy as one armed paper hangers trying to keep customers online during the event.
Executive director Darryl Swann tells CRN CCL was approached by a multinational company which had significant data centre assets in Christchurch gutted during the earthquake. Its backup tapes could not be restored.
“Our CCL engineers commenced an intense disaster recovery program on the storage by the dismantling the SAN- dripping wet from fire hoses and blackened from smoke – pulled each individual drive out of its casing, replaced the logic unit to the drives with one working unit, and connected each to CCL’s data centre infrastructure, where we wrote our own programs to simulate the SAN’s RAID set, in order to pull back all the data and getting them back up and running.”
Not only did CCL recover all lost data, but the client’s global systems were back up and running within 24 hours from a new location.
CCL’s DR provider is US company CommVault. Swann cited the company’s single console management for mail archiving, backups and compliance as being especially powerful. “This allows us to provide end-to-end IT services that are not just about data centres but just as much about creating robust and resilient solutions that feed into those data centres.”
Andy Purvis, Australian managing director of DR and back-up specialists Acronis cites a recent CIO survey which found a surprising lack of DR awareness amongst businesses in Australia and other leading economies.
In partnership with the Ponemon Institute - an independent analyst group specialising in data protection and security - Acronis surveyed over 4,000 IT practitioners across Australia, France, Germany, China, Japan, Brazil the UK and US.
One of the more startling findings was that less than half of respondents (45 percent) agreed or strongly agreed they had ample budget to support ‘comprehensive’ backup and DR operations. Only 43 percent of respondents said they conducted ‘file and backup’.
One of the big risks for companies that had experienced some sort of disaster is the problem of fading memory, which Gartner notes begins to kick in around nine to 12 months after a major event.
This finding was supported by the Acronis / Ponemon research which found only 25 percent of respondents were backing up more regularly following a natural disaster.
Interestingly though, respondents reported that human error was a large factor in downtime (54 percent), with power outages blamed 62 percent of the time.
“Backup and DR is a staple for all businesses and there is still opportunity for channel partners here,” says Purvis.
“These findings show there is still potential for the channel to educate business about the basics of DR, as well as take the opportunity that virtualsation and cloud present.”
Gartner estimates around 30 vendors share a business continuity and disaster recovery market. Defining the market is somewhat difficult as it is very broad, however Gartner says it grew by roughly a third last year.
It notes that server virtualisation has pushed down the cost of DR and back-up considerably making it a more viable consideration for companies that previously would not have had the budget.
“In recent years, the use of server virtualisation has reduced the server capital cost of IT disaster recovery management (IT DRM). Public cloud services are also becoming increasingly viable alternatives to more-traditional facilities’ recovery subscription and co-location services. Despite these cost reduction advances, however, one of the most significant IT DRM cost management challenges — reducing the cost of recovery plan testing and exercising — remains.
However, Gartner analyst and DR expert Roberta Witty stresses the need for companies to consider the cloud as another risk layer in their DR strategies.
“Cloud is another method of outsourcing your technology; it has the same kinds of risk as outsourcing does.”
The first priority, Witty says, is for companies to ensure their cloud provider is managing their recovery aspects in a way that satisfies the businesses’ own policies.
“You’re not outsourcing the governance of your data - you’re still responsible for it.”
Misunderstandings about data security and accountability within cloud contracts are increasingly leading to legal conflicts, especially in the US, Witty adds.
Veeam accidentally found itself a leading player in the DR space, by virtue of its fast growing installed base managing Microsoft and VMware virtual environments, explains the company’s local CTO Charles Clarke.
Founded in 2006, it has grown a strong market position helping organisations understand how to manage and back-up data within this new paradigm. Along the way it became apparent that companies had an opportunity to develop more effective DR management in virtual environments, provided the right tools are applied.
“DR is often touted as a benefit of virtualisation, especially because of its greater portability.”
“But without the right tools it can be a more onerous task than with traditional environments.”
“We’ve got to wrap our heads around the virtual environment. Disaster recovery is a benefit only if you understand virtualisation is not like the old way.”
Clarke added that 2012 saw a sharp rise in demand for storage as a service and back-up as a service as companies looked to cut costs.
“We want to offer offsite backups.”
He notes that high bandwidth costs and limited speeds have traditionally made offsite back-ups problematic for organisations. However, the improved capabilities around WAN acceleration are enabling them to more easily identify critical data and therefore reduce the network capacity needed to handle restorations from remote locations.
US-based systems management company Kaseya has been growing its DR business rapidly over the past few years, in line with the exponential rise in volumes of data organisations are now trying to manage.
“Because of the sheer volumes of data we’re amassing on a daily basis - how we’re addressing that market is changing, says the company’s A/NZ managing director Dermott McCann.
He tells CRN it helped several businesses affected by the Christchurch earthquake with its SaaS-based recovery solution. “We provisioned tenancies - helped them [NZ customers] run instances of their server on our platform.”
“The cloud gives the channel another arrow in their quivers.”
The growing use of cloud service providers - especially high amongst Australian businesses - is also creating opportunities for channel partners to add and manage the extra layer of monitoring organisations now need.
“Resellers need to provide another layer to monitor systems they don’t actually own. They have to monitor data in another cloud.”
He cites as an example, Salesforce.com, with which large numbers of Australian businesses now entrust large volumes of business critical information. Of course the BYOD trend presents a similar dilemma.
“There’s the risk of even less control yet the expectation is that apps will continue to be available,” says McCann.
Systems integrator and business continuity specialists SAI Global has seen a great deal of change in the DR space over the past several years. Over the years the company has developed specialised vertical competencies in order to properly service the requirements of customers, for instance in mortgage processing, insurance or real estate.
SAI’s general manager for IT, Peter Macarthur King explains a large part of the change has been due to the move to virtualisation.
He explains the company had been using a backup solution combining Veeam and IBM’s Tivoli but recently became frustrated with problems integrating the two products. That and the fact Tivoli is tape based. “Tivoli is way too cumbersome for our environment; it’s very tape oriented and doesn’t compress or deduplicate very well.”
“I couldn’t put hand on heart and say I could restore in two weeks.”
When Macurthur King joined SAI three years ago he was charged with completely overhauling the company’s tape-based back-up and DR strategy.
“We had total reliance on tape yet it was completely unreliable.”
“60 database backups failed in one night because Tivoli decided it didn’t want to do it anymore.”
Macarthur King began talks with new kid on the storage block, US company Actifio about using its virtualised back-up solution.
As Actifio’s founder and CEO Ash Ashutosh tells CRN, radical changes in the technology landscape the past several years, especially with the move to cloud and virtualisation, have demanded a new approach to DR and backup. One of the world’s foremost experts on storage, Ashutosh was formerly chief technologist and VP of storage with HP and is a successful VC based in the US. Actifio is on the war path to displace its larger rivals in the backup space and is seeking to recruit local partners to help it.
Japanese companies understand better than most the immense pain that natural disasters can inflict, especially since the devastating earthquake and tsunami, killing thousands and triggering one of the biggest nuclear emergencies in history.
Japanese imaging giant Ricoh has been quietly expanding its IT services arm (now worth $US3.1 billion) with a particular focus on DR. last year it snapped up local company IMC Communications with a view to bolstering its ability to help Australian companies better cope with extreme events.
Brought over from IMC, Matt Dixon is now head of IT services with Ricoh IT Services A/NZ. He notes that in years gone by there was an obvious emphasis on prioritising the importance of data and systems to avoid spending money on protection that wasn’t needed.
“No business can afford to have much downtime these days without it affecting their business and their brand.”
At the same time, the competitive needs of “real time” businesses such as banks, traders and even online retailers have raised the bar both in terms of technological requirements and the need for more rigorous planning.
“It’s all about risk profile and how damaging every second of downtime is to the business.”
“There’s different levels of dependence on IT systems.”
The most impressive asset in the company’s arsenal it is long term lease of the data centre underneath Sydney Olympic (now ANZ) Stadium at Homebush, first built to handle the year 2000 games and for many years retained by IBM and Telstra. It’s now the main DR centre for Ricoh IT Services in Australia, and also operates as a co-lo facility for other businesses.
Ricoh IT services is currently deploying microwave services and dark fibre to connect the centre with its facilities in Eastern Creek, Frenchs Forrest and offices in North Sydney.
Dixon notes that companies must consider DR as a critical part of their wider business continuity planning.
“It starts with a design and review of what needs to be replicated, but it’s definitely not a set and forget.”
“It has to be constantly considered in all changes to the production environment.”
Craig Allen, from Sydney-based cloud recovery specialists Offis, notes there are essentially three types of disaster recovery customer, although the groups are not mutually exclusive.
The first and most obvious are organisations that have large amounts of sensitive data, such as financial services companies, where systems failure and business failure are the same thing. Then there’s companies that have been burnt and have a ‘never again’ mentality. The third group are companies that are yet to understand the importance of DR altogether.
“You have to align customers’ expectations with the technologies,” Allen says “For us, the question will always be ‘how long can it be down for?’”
“If you’re working at C-Level they’re thinking ‘what’s the cost to the business and how do I mitigate against it?’”
Offis works with a number of financial services organisations which typically have very tight RTO (recovery time objective) and RPO (recovery point objective) requirements.
But with most other clients the company strives to understand what aspects of the business are more critical than others in order to develop the most practical and cost effective solution. Key to this, Allen says is developing a clear understanding of the client’s core applications.
“You must be application aware. You can’t just take any app, snapshot it, move it to another location and turn it on.”
He cites a recent example of a toll booth application recently deployed in London. Due to an oversight at the planning stage, drivers were prevented from leaving the facility after 2am, leading to pandemonium.
“It’s very hard without extensive in-depth scoping to identify every possible risk.
Companies need to familiarise themselves with SAN (storage area network) and WAN (wide area network) replication solutions.
Allen echoes Gartner’s observation that the DR space is yet to fully blossom into a market in its own right, with many businesses reliant on the inbuilt capabilities of their core platforms.
“DR is typically offered as an extension to existing services.”
Previously this might have been cited as evidence for a lack of maturity in the DR space, now it’s seen as further proof of the impact of cloud computing and virtualisation on business.
Offis boasts of having been a cloud solutions provider since its formation in 1997 and helping its customers achieve faster, cheaper and more reliable DR. For instance, its mission critical clients are serviced via an infrastructure as a service solution which guarantees 100 percent up-time. Real-time compression of Offis’ IBM V7000 storage infrastructure yielded it a 48 percent performance improvement with savings of $40,000.
Acronis’s Purvis notes a major factor in growing awareness around DR, especially amongst SMBs, is fear about the safety of personal data, be it family photos, music and the like.
It’s this change Purvis says explains the company’s transition to a business backup provider as SMB managers saw the company’s consumer offerings as a cheap and viable business backup option.
“We started to look at our customer lists and started to see big names in there,” Purvis says.
Another factor helping to drive uptake was the fact that individuals with organisational silos were buying Acronis behind their IT managers’ backs rather than risk delays due to onerous testing and approvals processes.
“We find that departments within organisations still have to solve problems - going through IT department tends to be too slow.”
The fact is reflective of the need for more coordinated, company-wide approaches to disaster recovery including the development of proper policies and procedures.
Purvis explains that Acronis has a unique channel model whereby it invites companies to download a free trial of its solutions online. Once a customer makes contact it then refers the lead to one of its partners
“We work with channel where they are responsible for all components of the solution - but we provide the pipeline to help them close.”
Getting budget for DR remains a challenge for IT managers.
“It’s one of those things where people say ‘if what we’ve got is good enough we’ll live with it’.”
It’s common, he adds, for companies to do only “incremental” back-ups. “But if they have a server with terabytes of data, it could take three days to recover.”
“Companies need to work out whether DR is a strategic imperative or is it tactical.”
“In the enterprise space people know about this but in the SMB world they’re stuck in this middle ground of ‘we think we’re backing everything up and have everything we need’.”
Purvis recently attended an industry event where he met with 15 Australian CIOs, and only two could say who their main DR vendor was.
He adds the BYOD phenomenon highlighted the need for companies to urgently address their current DR plans. “The mobility space is certainly a big frontier in terms of risk.”
Increasing levels of risk for many companies is the high degree of complacency about the mobility trend, Purvis explains. He likens the situation to conversations companies had with mainframe administrators in the 80s about the imminent flood of PCs. “They weren’t interested.”
“The tablet and smartphone device revolution is creating a whole new landscape of opportunity or risk depending on how you approach it,” Purvis says. In the US for instance, it is estimated that a smartphone is lost or stolen every 56 seconds. That’s a lot of business contacts and communications.
One of the more recent notable developments within the DR space is recovery assurance, which concerns the potential for today’s technology to support highly sophisticated DR testing solutions. Gartner for its part advocates the building of virtual test centres.
“IT administrators can automate simulated failover scenarios to the recovery site as often as desired,”
it says. “IT administrators can carry out recovery tests of individual IT services weekly or daily, rather than yearly or quarterly. Reporting capabilities allow IT administrators to be alerted to potential problems based on either standard or user-defined thresholds.”
Gartner notes recovery assurance is a relatively immature category for DR, with only a handful of supporting products and vendors.
“However, as the deployment scope of virtual machine recovery increases, so too will the need for recovery automation functionality to move toward “guaranteed” recovery times and reduced recovery plan testing and exercising costs.”