Cyberinfrastructure in New York State - Grid Computing: Infrastructure, Service, and Applications - page 45

parameters for the processing of some representative XML-based grid
documents on multicore nodes (Bhowmik et al., 2007).

2.3.9 Protected Grid Data Transfer

Widely available and utilized grid servers are vulnerable to a variety of
threats from denial-of-service attacks, overloading caused by l ash crowds,
and compromised client machines. The focus of our work is the design,
implementation, and evaluation of a set of admission control policies that
permit the server to maintain sustained throughput to legitimate clients
even in the face of such overloads and attacks. We have developed several
schemes to effectively, and importantly in an automated fashion, deal
with these attacks and overloads. We have implemented these schemes
efi ciently on an active network adapter-based gateway that controls access
to a pool of backend data servers. Performance tests conducted on a sys-
tem based on a dual-ported active NIC demonstrate that efi cient optimi-
zation schemes can be implemented on such a gateway to minimize the
grid service response time and to improve server throughputs under
heavy loads and denial-of-service attacks. Our results, using the GridFTP
server available with Globus Toolkit 4.0.1, demonstrate that even in adverse
load conditions, the response times can be maintained at a level similar to
normal, low-load conditions (Demir et al., 2005, 2007).

2.4
In this section, we present the timeline and chain of events that led to the
establishment of a grassroots New York State Cyberinfrastructure initia-
tive. We also discuss the relationship between this grassroots initiative,
the NYS Grid, and Miller's Cyberinfrastructure Laboratory.

New York State Cyberinfrastructure Initiative

2.4.1 Timeline

In the early 1990s, Miller's research group began to use commercial paral-
lel computers in an effort to solve the phase problem of crystallography,
given atomic resolution X-ray diffraction data. We used a variety of such
machines, including a Thinking Machines Inc. CM-2 and CM-5, an Intel
iPSC/2, an Encore Multimax, and various other early commercial parallel
machines. Simultaneously, Miller's group used a laboratory of Sun work-
stations and ran their Shake-and-Bake method of structure determination
using the RPC (remote procedure call) protocol to employ a master/worker
solution of Shake-and-Bake. This was similar to a CONDOR-based
approach of stealing backgrou nd cycles to perform computationally inten-
sive tasks that can be broken up into discrete and loosely coupled tasks.