If you’ve ever been left twiddling your thumbs as you wait for your webmail service to present a usable interface or your SaaS spreadsheet to become responsive, it’s likely that the networks that lie between you and your application provider are responsible. Data surges from its home network at high speed before being bogged down somewhere on its way to you. The further away from the source you are, the longer you’ll be waiting.
Network Latencies Kill Application Performance
Movies and TV shows often represent the internet as an idealized cloud of bright pinpoint signals whizzing around the world in straight lines at the speed of light. There’s almost no delay between sending and receiving a packet of data. The reality is much different. Data sent from and to cloud platforms has to contend with a tangled mess of Internet service providers, interconnects, routers, switches, copper wires, fiber, broken connections, dropped packets, overloaded networks, and the complex miscellany of networking technology that it encounters on its trek across the planet.
The business models of Internet Service Providers and other bandwidth providers are frequently built around SLAs that focus on data delivery, but not on the performance of their networks or of the applications that run over those networks. Application performance is not a core consideration. Underperforming networks are outside of the control of cloud providers, and the performance issues they create have the potential to seriously impact adoption of software-as-a-service applications among consumers and in the enterprise.
“Slight changes in network performance may have significant consequences to application performance. Applications and technologies such as video, VoIP, and virtual desktop infrastructure (VDI), as well as different flavors of unified communication (UC) technologies, are increasing the complexity of wide area network (WAN) traffic and are posing new challenges for managing WAN performance.”
None of which is good news for application providers. Enterprise users depend on fast and consistent performance from their applications and poor performance can impact adoption and increase customer attrition rates.
Building A Global Application Platform
Historically, the obvious solution to this problem — reducing the geographic and network distance between customers and application servers — has been fraught with complexity. For applications with a global user base, the optimal solution is a platform of geographically distributed application servers, each providing service to a limited area. But, deploying multiple application servers with multiple vendors is a logistical and managerial nightmare. Or at least it is without the services of a cloud integration layer and cloud marketplace like ComputeNext.
ComputeNext provides an easy to use interface for comparing cloud vendors. Infrastructure can be deployed and managed from within ComputeNext, and because there’s a single point of payment for all of ComputeNext’s cloud partners, cost management isn’t nearly the problem it would be if application developers were to contract with each cloud vendor individually.
Businesses are enthusiastic about cloud applications for all the reasons we’re familiar with: cost, convenience, and flexibility; but without a geographically distributed application hosting platform, performance problems will harm adoption rates.