Geoscience Reference
In-Depth Information
Fig. 4 Experimental system with thick client interface
than 1 % of the service time. The server was hosted on an Intel Core i7 2670qm
running at 2.2 Ghz, 4GB DDR3 665 MHz, Windows 7 professional sp1 64bit. The
database server (MSSQL 2008 R2) was running on the same host as the Weda
server so its latency is included in total amount of RPT. It was found that RPT times
mainly consist of latency of data access layer (99 %). The test case for measuring
response time instability has been de
ned with constant payload of GetCapabilities
operation invoked at OGC SOS webservice. Every 10 s for 3 h a request was sent
and results were measured to give us more than 1,000 samples. The server pro-
cessed each request by proper serialization at each layer up to the bottom data
access layer. Backward propagation of results was packed into response frames by
Weda API and metadata about server processing times was glued into the response.
An illustration of setup and measuring points can be seen in Fig. 5 .
RTT includes a time for request forwarding achieved by our reverse proxy. This
was used to simulate such a device
is delay. The Weda architectural style is a hybrid
architectural style that we have derived from other network-based standards, such as
web services [ 5 ] and HTML5 web-sockets [ 6 ] to get a practical real-time SOA 2.0 [ 7 ].
'
Fig. 5 Benchmark setup
Search WWH ::




Custom Search