Minimizing protocol processing in multimedia servers - implementation and evaluation of network level framing

Abstract

Data servers for multimedia applications like news-on-demand represent a severe bottleneck, because a potentially very high number of users concurrently retrieve data with high data rates. In the Intermediate Storage Node Concept (INSTANCE) project, we develop a new architecture for media-on-demand servers that maximizes the number of concurrent clients a single server can support. Traditional bottlenecks, like copy operations, multiple copies of the same data element in main memory, and checksum calculation in communication protocols are avoided by applying three orthogonal techniques: network level framing (NLF), zero-copy-one-copy memory architecture, and integrated error management. In this paper, we describe how to minimize the transport level protocol processing using NLF. In particular, we look at how NLF is implemented, and we present performance measurements indicating a large performance gain. The protocol execution is minimized to about 450 cycles per packet regardless of packet size, i.e., a reduction of about 87% compared to 1 KB packets and more using larger packets. Consequently, the total server-side processing overhead is decreased by at least 50%.