Increased latency, often with congestion, leading to sluggish, unpredictable application performance for branch-based users can undermine cloud computing efforts.
Backhauling all Internet traffic though HQ or data center makes network management and security simpler, at the expense of application access performance.
Videoconferencing, large video files, enterprise synchronization, public cloud computing all will contribute to the need for a lot more bandwidth
BYOD cloud synching, SaaS, cloud computing, video just a few of the reasons
Cloud services, SaaS, the power of the colo, Internet economics, MPLS expense, ensuring high performance, predictability and reliability for WAN users, and visibility and control for network managers all key.
WAN Virtualization enables reliable, predictable high-performance in the face of packet loss on a network path.
Using a multi-segment architecture combined with a dedicated core can drastically reduce the impact of WAN packet loss on TCP applications.
Handling loss differently, mitigating the effects and hiding the loss from end stations can improve WAN application performance.
Multiple approaches are possible, and drastically reducing the number of packets that need to traverse the WAN is one good one
TCP's primary mechanism for signaling congestion causes application performance degradation in many ways, and was never optimized for high-bandwidth WANs or interactive applications over the WAN.
