Dell H730P Manual do Utilizador Página 5

  • Descarregar
  • Adicionar aos meus manuais
  • Imprimir
  • Página
    / 18
  • Índice
  • MARCADORES
  • Avaliado. / 5. Com base em avaliações de clientes
Vista de página 4
A Principled Technologies test report 5
Database performance and resiliency in the Intel processor-powered
Dell PowerEdge R630 running Microsoft SQL Server 2014
Scenario
HDD - 4 DB
(1 SQL instance)
SSD - 4 DB
(1 SQL instance)
SSD - 8 DB
(2 SQL instances)
SSD - 24 DB
(4 SQL instances)
Client 1
4,719
46,627
29,443
10,642
Client 2
7,376
46,597
30,184
10,614
Client 3
7,628
46,819
30,283
10,623
Client 4
4,362
46,561
30,325
10,596
Client 5
30,517
10,678
Client 6
30,722
10,466
Client 7
30,711
10,567
Client 8
30,080
10,320
Client 9
10,611
Client 10
10,580
Client 11
10,565
Client 12
10,407
Client 13
10,528
Client 14
10,599
Client 15
10,550
Client 16
10,638
Client 17
10,631
Client 18
10,360
Client 19
10,578
Client 20
10,570
Client 21
10,393
Client 22
10,627
Client 23
10,655
Client 24
10,516
Total OPM
24,085
186,604
242,265
253,314
Variability
29.1%
0.6%
4.2%
3.4%
Figure 4: Performance and variability for SSD scenarios we tested.
Another factor in application service quality is the length of the delay between
the user making a request and receiving the requested data. This delay, referred to as
latency or response time, is measured in milliseconds, and is another useful metric for
showing the performance difference between the SSD- and HDD-based solutions. We
used the average of return trip (total) over all orders in milliseconds, called
rt_tot_avg_msec counter, in the DVD Store 2.1 benchmark to determine the average
round-trip latency for the test run on a sample client. In our testing, we found that the
SSD-based solution was almost 60 percent lower in response time as reported by the
benchmark tool than the HDD-based solution (see Figure 5).
Vista de página 4
1 2 3 4 5 6 7 8 9 10 ... 17 18

Comentários a estes Manuais

Sem comentários