Warehouse-Scale Computers to Exploit Request-Level and Data-Level Parallelism - Computer Architecture: A Quantitative Approach - page 488

FIGURE 6.27 Utilization distributions across cluster, without and with consolidation .
6.31 [10/Discussion] <6.2, 6.4, 6.6> System-Level Energy Proportionality Trends : Consider the fol-
lowing breakdowns of the power consumption of a server:
CPU, 50%; memory, 23%; disks, 11%; networking/other, 16%
CPU, 33%; memory, 30%; disks, 10%; networking/other, 27%
a. [10] Assume a dynamic power range of 3.0× for the CPU (i.e., the power consumption
of the CPU at idle is one-third that of its power consumption at peak). Assume that
the dynamic range of the memory systems, disks, and the networking/other categories
above are respectively 2.0×, 1.3×, and 1.2×. What is the overall dynamic range for the
total system for the two cases?
FIGURE 6.28 Overview of data center tier classifications . (Adapted from Pitt Turner IV et
al. [2008].)
b. [Discussion/10] What can you learn from the results of part (a)? How would we
achieve beter energy proportionality at the system level? ( Hint : Energy proportional-
ity at a system level cannot be achieved through CPU optimizations alone, but instead
requires improvement across all components.)
6.32 [30] <6.4> Pit Turner IV et al. [2008] presented a good overview of datacenter tier clas-
siications. Tier classifications define site infrastructure performance. For simplicity, con-
sider the key differences as shown in Figure 6.25 (adapted from Pit Turner IV et al. [2008]).
Using the TCO model in the case study, compare the cost implications of the different tiers
shown.
6.33 [Discussion] <6.4> Based on the observations in Figure 6.13 , what can you say qualitat-
ively about the trade-offs between revenue loss from downtime and costs incurred for up-
time?