Storage: Enterprise Systems Become More, Not Less, Confusing

This is the year that people start building storage hardware from off-the-shelf, high-performance flash storage devices. This is the problem that NetApp and EMC tackled with spinning disks in the ’90s. But flash devices are so fast that the software architectures invented back then don’t work anymore. Flash has turned all the other components in your storage system—CPU, memory and network—into performance bottlenecks. Storage companies need to think about system architecture if they want to expose the available performance from high-performance flash. —Andrew Warfield, CTO of Convergent.io

Servers are sprouting onboard 10GB network interface cards (NICs), and 10GB switching has become affordable and is giving Fibre Channel and Infiniband an existential crisis. Large-scale cloud providers have demonstrated that they know how to build software that gets the performance and durability that data center administrators want from enterprise storage, at scale, without an expensive, packaged array. In short, technology is really messing with the definition of enterprise storage right now. —Andrew Warfield, CTO of Convergent.io

Cloud providers will place high importance on delivering offerings suitable for mission-critical enterprise applications, to shift away from the typical test/develop deployment scenario. As more enterprises look to move their legacy applications to the cloud, they’ll find a hybrid cloud to be the most suitable for their specific needs. It will combine the scalability and savings of a public cloud with the security of a private cloud for compliance requirements—such as Peripheral Component Interconnect (PCI), the National Institute of Standards and Technology (NIST), the Federal Information Security Act (FISMA) and the Health Insurance Portability and Accountability Act (HIPAA)—and enable the isolation of sensitive data. —Simon Aspinall, chief vertical markets, strategy and marketing officer at Virtustream

Detection as a mechanism for protection is moving toward extinction. This methodology must expire in 2013 as it is plainly obvious that the new generation of malware is becoming undetectable. Enterprises that try to rely on technologies that detect malware as the mechanism for remediation are doomed to failure. These types of attacks can remain dormant for months before reaching their intended targets, and removing them after they strike may be pointless: The damage has already been done.—Tal Klein, senior director of products at Bromium

Advanced persistent threat (APT) creators are developing versatile components—which can shift from one IT platform to another, akin to the Avian and Swine flu that can jump to different species. We will see malware that begins life on one OS and then hops onto another. Synchronizing one device to another has already transcended the typical use for data backup. As applications move beyond “living” on a single device, the likelihood of targeted malware that takes advantage of data synchronization to move from unprivileged devices onto privileged ones becomes very real. In a post-Stuxnet world where malware propagates from Windows PC laptops to Siemens S7-300 manufacturing control systems through exploits in the controller application, it isn’t a big leap to imagine that malware can propagate from phone to laptop and laptop to tablet through operating system and application vulnerabilities. —Tal Klein, senior director of products at Bromium

Corporations traditionally have been able to control the timing of new OS deployments because they control the purchase of PCs and their operating systems. But what happens in a BYOD world where employees buy their PCs directly from Best Buy and Fry’s? Those PCs are loaded with Windows 8, and they are coming to an enterprise near you. We expect to see a race to deploy virtual desktop infrastructure (VDI) as IT’s way of maintaining control over the timing of Windows 8 support and for migrating legacy applications to Windows 8. —Lee Caswell, founder and chief strategy officer at Pivot3

Another factor that will push the adoption of VDI is that, as one industry pundit commented, 2013 is the year that BYOD becomes BYODDDDD. Device sprawl is upon us. The latest IDC forecasts predict 40 percent unit growth in year-over-year tablet shipments to 165 million units in 2013. More devices, particularly multiple devices per user, present huge management challenges for IT when users expect to access corporate applications securely from any device at any time. VDI is one of the few tools available to soothe the pain and security risk of device explosion. —Lee Caswell, founder and chief strategy officer at Pivot3

IT groups and budgets aren’t expanding to match the growth in devices. In fact, it’s quite the opposite. IT budgets are being held flat at best. In this environment, 2013 will see IT investing in VDI to regain the leverage of centralized infrastructure, security and backup investments for the disparate set of user-owned devices. It’s the tried and true IT model that will hold true for device proliferation as surely as PC proliferation drove investment in Novell NetWare LAN adoption in the 1980s. —Lee Caswell, founder and chief strategy officer at Pivot3

For greenfield infrastructure projects—particularly public cloud, private cloud, infrastructure for software as a service (SaaS) hosting and big data farms—we should expect open-source software or open-source-compatible software to win the bulk of the deals. Proprietary infrastructure will not disappear in 2013, but it will have a great deal of trouble gaining new ground and winning new projects. —Jay Judkowitz, director of product management at Nimbula

Organizations will seek to get more value from the infrastructure investment in internal clouds. To obtain ROI from virtualization and cloud adoption, companies will look to productivity improvements to reduce operating expenses and improve economics. Changes in the technology, people and processes to streamline IT operations may entail creating more fluidity between functional groups, increasing the administrator-to-server ratio through better management tools that allow higher densities of workloads per host, and automating processes wherever possible. —Shmuel Kliger, co-founder and CTO at VMTurbo