This article first appeared in the Summer 2017 issue of The Record.
A recent IBM study found that 90% of all the data in the world was created in the last two years. What’s more, according to Gartner, data volume is set to grow by 800% over the next five years, 80% of which will be unstructured data that will require more IT engineering and hardware resources. According to Jimmy Tam, CEO at Peer Software, this has led to huge challenges when it comes to file management.
“Our customers are drowning in data and, as a result, are being held back,” he explains. “One of our customers, HKS Architects, said that ‘with the volume of data growing by up to 10% each month, information management has become a burden’. Another customer has expressed concern about the diversity of the storage platforms they managed. They had Windows File Servers at branch offices; NetApp cDOT controllers at their main data centres; as well as EMC Isilon storage.”
According to Tam, these issues are just the tip of the iceberg. “We’re also seeing challenges around the management complexity of moving data workloads to the right storage platform and to the right location in a multi-site, multi-vendor environment. Then there’s the inability to analyse the data being created across the organisation to drive real business intelligence,” he says.
It’s clear, then, that something needs to change – and this is front of mind for Tam. “We have created Peer Global File Service (PeerGFS) to help storage administrators handle the challenges they face every day,” he says.
PeerGFS helps users in several ways, not least by providing mixed storage support. “We understand that storage administrators today normally have Windows File Servers at the branch office and large network attached storage or a storage area network at their major data centres. In addition, they want to mix in both private and public cloud storage, depending on their requirements,” Tam says. “Historically, each of these platforms would create data silos where information generated and stored on one would be difficult to move across to the others. However, due to our partnership and API integration with major storage platform vendors such as Microsoft, NetApp and EMC, PeerGFS enables our customers to choose the right solution for each different workload without worrying about vendor lock-in and information silos. PeerGFS enables the seamless replication and movement of data across a mixed hybrid cloud environment.”
PeerGFS also helps companies to manage the different types of data they hold. “We always talk to customers about hot, warm, and cold data,” Tam explains. “Hot data is data that is being actively worked on today and generally has been accessed or modified within the last six months. Hot data needs to be immediately accessible by employees for maximum productivity and so should reside locally at both the branch office and head office data centres. Warm data is that which hasn’t been modified within the last 6-12 or 6-18 months, but is still being read/accessed on an occasional basis. This generally will only reside at the head office data centre. Cold data is that which has not been modified and accessed in over 12-18 months and generally should be moved off to slower, denser, cheaper disks or to the cloud.
“PeerGFS enables the real-time bi-directional replication of hot data across the branch and head office data centres so that a local copy of hot data resides everywhere,” Tam continues. “Version conflicts across the active-active distributed data centres are prevented via integrated file locking. That means that when a file is being edited at any one site, the other hot data copies are locked down for read-only access until file edits are saved and replicated or updated to the other sites.”
Integrated continuous data protection and high availability are also an inherent part of the solution. “PeerGFS creates exact folder and file replicas across mirrored data centres,” explains Tam. “And with Microsoft DFS Namespace support, customers also rely on us for their disk-to-disk data protection and high availability requirements.”
What’s more, because PeerGFS is integrated with the cloud, offsite backup storage requirements that were traditionally done via tape can now be replaced with public cloud storage from Microsoft Azure. “Through Microsoft Azure integration, customers can expand their disk-to-disk data protection to disk-to-disk-to-cloud with a backup snapshot catalogue provided by PeerGFS for volume, folder or file level point in time restore capabilities,” Tam says.
Finally, PeerGFS provides ‘analytics ready’ backups in Microsoft Azure. “Most backup solutions on the market today store snapshots in the public cloud in proprietary formats,” explains Tam. “As such, the data that resides inside the snapshot is not immediately available to Microsoft Azure AI, machine learning or other cognitive services without a cumbersome and costly data extraction and transformation process. PeerGFS is different in that each file being replicated to Microsoft Azure is created as a separate object so that it is immediately useable by any of Microsoft Azure’s analytics engines to drive real-time, efficient business intelligence.”
As data continues to grow, Tam expects that file management will quickly evolve. “Customers will need solutions that analyse data as it is being created, automatically place it on the right storage platform based on the usage trend of such data and the associated capabilities required of the data workload, and ultimately also move the data across its lifecycle along the appropriate storage tiers (hot, warm, and cold),” he concludes. “As the mountain of cold, backup data grows, the ability to search, find and retrieve old information for compliance and other purposes must keep pace.”