September is always a busy month for mobile news, this year highlighted by impressive new iPhone and Galaxy Note releases, an iOS update and even an unorthodox new Blackberry phone. These events have led me to ponder the use of mobile devices, whether smarthphones or tablets, in business where these now-indispensable consumer items (e.g. more people carry a smartphone than wear a watch) are still supplementary and subordinate to PCs in the workplace. The question is why? Why have smartphones – indeed, the mobile device market is utterly dominated by phones, more so than ever with the rise of large-screen phablets – become the preferred and often exclusive information and communication platform for younger people yet often remain accessories at work? In a recent column, I attempt to answer these questions and provide guidance for business professionals thinking about turning their iPhone or Galaxy into the primary work platform and solo travel companion — and yes, it is actually possible, I cite an existence proof.
Mobile devices aren’t just glorified typewriters.
Smartphones have evolved into the do everything Swiss Army information appliance most people are never leave home without, yet they remain glorified PDAs for most business users. Why the dichotomy? Are mobile devices and their all-important apps inherently optimized for lightweight, ephemeral consumer needs and thus unsuitable for the heavy lifting of memo writing, spreadsheet analyzing and database mining of business? Or is this a case of the mobile business ecosystem following the early money and large market into consumer-oriented features and applications with business needs relegated to a second wave of mobile development? I think both hypotheses are true, but the net is that adapting work styles, processes and software to mobile devices presents an enormous business opportunity.
In the column, I explain why, in an era of smartphones with PC-class performance, ever-present high-speed wireless networks and rich backend cloud services, there’s no reason why business execs must remain tethered to the evolutionary descendent of typewriters.
Intrigued by the concept of a converged infrastructure but worry you lack the expertise to DIY? Big IT vendors see big business building insta-cloud infrastructure
IW Report Cover
For IT organizations wondering how to actually build a private cloud, my latest report is for you. Plenty of CIOs like the idea of a converged data center but get headaches thinking about the complexity. Even given wide support for open standards, stitching discrete server and storage systems into a software-abstracted cloud is not for the faint of heart. Enter preassembled private cloud stacks, such as Dell’s PowerEdge C-series, Hewlett-Packard’s ProLiant SL Scalable Systems, IBM’s NeXtScale System, Nutanix and now VMware with its EVO RAIL reference design. These systems, which either bundle or are qualified on popular cloud stacks, can cost-effectively deliver scale-out cloud functionality, high density, and manageability without significant compromise on features.
VMware EVO RAIL chassis
As my summary article points out in the full report downloadable here, InformationWeek survey data shows CIOs and IT leaders have the will to build private and hybrid clouds, it’s the execution that’s problematic for many. One reason is that there is no standard blueprint for building a next-generation software-defined infrastructure.
EVO RAIL server module
The term “private cloud” describes everything from automated virtual machine system administration to fully orchestrated and self-service compute, storage, and network resources that end users and developers can provision themselves. Given this uncertainty and complexity, the pioneers tend to be businesses that began on public cloud infrastructure and are comfortable living on the bleeding edge of technology.
Download the full report for some concrete recommendations, but the choice of insta-cloud platforms will largely depend on your existing virtualization environment, desire for cloud stack heterogeneity and openness, use of public IaaS services like AWS, Google Cloud or Microsoft Azure, and whether branch office servers and remote users are in the mix.
IDF, Intel’s festival for hardware geeks, better known as IDF, was largely overshadowed by some announcements from that not insignificant fruit company from Cupertino. The coincident timing was unfortunate since Intel provided plenty of news for both PC watchers and data center designers. On the server front, the company released a major update to the Xeon line based on the Haswell microarchitecture and as I discuss in this column, the new products includes several enhancements specifically designed to improve performance with virtualized workloads and multi-tenant cloud deployments.
Xeon E5 v3 package
One of the notable changes addresses the bane of many cloud customers: noisy neighbors. Disruptive neighbors are a major problem in any shared environment, whether it’s a cubicle farm, apartment building or cloud service. However unlike the party animal upstairs who regularly keeps you up until the wee hours, in multi-tenant cloud, IaaS environments, you typically don’t know precisely who the noisy tenant is. The only thing for certain is that at least one workload on the same physical server is hogging resources, slowing down everyone else. But wait, isn’t virtualization supposed to solve this? Doesn’t each workload have its own virtual slice of the system, walled off and isolated from every other VM? Of course; but the dirty little secret of virtualization is that the software abstraction only goes so far and there remain several shared resources on every system.
Xeon E5 v3 die
As I detail in the column, the most visible resources hogged by noisy neighbors are network and disk I/O, however these are easy to spot. However another shared element, processor cache memory, has been completely invisible and inaccessible to the hypervisors powering all cloud services: at least until now. Intel, in its new Haswell architecture Xeon E5 v3 processors, just announced at IDF, takes a first step at silencing noisy neighbors at the source, namely the on chip execution environment. Click through for the rest of the story.
I take an in-depth look at IoT network design in a new InformationWeek Report (available here, registration required), but the key takeaways in this summary column are that the Internet of Things demands reliable connectivity, yet standards are up in the air.
InformationWeek IoT Tech Digest
While consumers and Wall Street analysts tend to conflate “IoT” with “wearables tethered to smartphones,” most enterprises don’t consider mobile devices with 3G/LTE connectivity to be the most important — or most common — edge devices. There’s a big difference between a fitness band tapping a health-tracking app and an industrial control system managing HVAC and building security cameras on a few dozen sites. While consumers eagerly await a svelte new iWatch or Android wearable, for organizations such as Union Pacific, FedEx, GE, and ConocoPhillips, the IoT is here.
As I point out, the first thing to keep in mind when mobilizing your IoT plan is the distinction between consumer and industrial. IoT consumer devices will likely either be tethered to a smartphone with a vast array of communications options and local storage and processing, or they will be used in controlled environments with access to WiFi and power, either a wall socket or easily replaceable or rechargeable batteries. The same can’t be said for most industrial applications — like pollution monitors affixed to traffic signs or water sensors in hard-to-access pipes, both of which run on a battery for months or years between servicing.
Intel’s IoT gateway architecture
IoT distributed network topology
Source: Chuck Hollis, VMware
IoT networks need to be more flexible than the typical hub-and-spoke or leaf-spine designs IT is familiar with. Options include using intelligent gateways and edge mesh networks. Download the full report for details.
One (of several) epiphanies I had at VMworld occurred during an extended discussion with Jeetu Patel, GM of EMC’s Syncplicity business unit, about the importance the company places on its mobile app usability and design. The revelation was that a company better known for its big storage iron could nurture a software division that can teach many Silicon Valley startups a few things about app design, usability and quality. As I write in this column, mobile devices have created a renaissance for user interface designers that are simultaneously trying to exploit their attributes of tactile, high-resolution screens accessible in any situation, while mitigating their drawbacks of limited screen real estate, lack of a physical keyboard and a single app per screen focus. Together, these create both opportunities and challenges for application developers. With at least 1.2 million titles on Apple’s App Store having been collectively downloaded over 75 billion times, and Google’s Android Play Store reporting nearly identical numbers, it’s clear mobile apps are big business with a huge audience.
Syncplicity iPhone tab structure.
As I discuss in detail, while Apple and a few standout developers may be best known for elegant apps, few people would associate a division of EMC, the storage hardware provider to enterprise IT, more known for its big storage arrays than mobile app finesse, as one of today’s most innovative mobile developers. Yet give the company’s Syncplicity title a spin and tell me it’s not one of the slickest, most well-designed, intuitive and beautiful apps you’ve seen.
Synplicity home screen, parallax effects.
As I wrote in this column on the eve of VMworld, the last week of August saw VMware hold its annual festival to virtualization, cloud and the myriad ways software is eating the data center, for an event that’s become a milestone on the annual tech calendar where IT managers and sysadmins rub shoulders with industry executives, analysts and entrepreneurs. As always, the company used the huge event with over 22,000 attendees and worldwide media coverage to showcase new software, services and partnerships, but the overarching theme was clear: VMware will not be left behind as the cloud dramatically reshapes IT strategies and operations. The company continued its evolution from a purveyor of hypervisors and systems management software to a provider of a full spectrum of cloud infrastructure software and services by opening with the expected enhancements to its core vCloud management software and NSX virtual networking product. It went on to announce a marketing exercise designed to simplify branding and sales by bundling and renaming several management products into the new vRealize Suite and rechristening its vCloud hybrid service vCloud Air.
EVO RAIL server module
Source: author photo
All are the sorts of technology upgrades and marketing tweaks one expects from a company’s biggest trade show, however these merely set the table for the big news: VMware has developed its own OpenStack distribution running on native vCenter infrastructure and using its ESXi hypervisor. Read the rest of my analysis in my Forbes column.
VMware also unveiled a new hyper-converged server reference platform, EVO RAIL, and had several OEM partners including Dell and Fujitsu showing off systems on the VMworld show floor. I’ll have more to say about EVO RAIL in an upcoming InformationWeek Report on private cloud infrastructure, but it EVO has all the look of a big software company, VMware, attempting to commodify hardware around a standard and undifferentiated platform, while its cloud software takes center stage.
EVO RAIL chassis