High-Efficiency Remote Cloud Data Center Backup with Intelligent Parameter Adaptation - Technologies and Applications of Artificial Intelligence - page 31

5 Experimental Results and Discussion

This section will go for the remote data center backup, the stress test, as well as the
evaluation of total cost of ownership and performance index among various data
centers. Finally, the assessment about the effectiveness and efficiency among various
data centers have done well based on cost-performance ratio. All of tests have
performed on IBM X3650 Server and IBM BladeCenter. The copyrights of several
databases applied in this paper have Apache HBase and Apache Cassandra are of
NoSQL database proposed this paper, but otherwise Cloudera HBase, DataStax
Cassandra, and Oracle MySQL are alternative databases.

5.1 Intelligent Parameter Adaptation
In order to improve the quality of services (QoS) on remote data center backup, we
have to deal with the crucial problems about jitter, loss, latency, and throughput in
this experiment. According to the diagram as shown in Figs. 1 and 2, an intelligent
adaptation using an adaptive network-based fuzzy inference system (ANFIS) with
particle swarm optimization (PSO) approach is employed, which have utilized
the CPU clock rate, the size of SDRAM, NIC bandwidth, message, the ratio of read
to write operation per second or the number of read plus write operation per
second, and the number of messages per second, respectively, as listed in Tables 1
and 2, to adjust two parameters (1) hbase.regionserver.msginterval and (2)
hbase.ipc.server.callqueue.read.ratio in HBase and three parameters (1)
commitlog_total_space_in_mb, (2) compaction_throughput_mb_per_sec, and (3)
reduce_cache_capacity_to in Cassendra, as shown below in Figs 3 and 4.

Fig. 1. Intelligent parameter adaptation for
remote HBase backup

Fig. 2. Intelligent parameter adaptation for
remote Cassandra backup

Table 1. Five operational mode for HBase backup

Size of
DRAM
(GByte)

Read/Write
Ratio (per
sec)

Operational
Mode

CPU Clock
Rate (MHZ)

Band Width
(Mbps)

Message
(per sec)

1 2400 32 77.79 0.9 15

2 2300 32 77.79 0.5 5

3 2200 32 77.79 0.7 9

4 2100 32 77.79 0.8 12

5 2000 32 77.79 0.6 7