Mode of access: World Wide Web.

System requirements: Adobe Acrobat Reader.

Part of: Synthesis digital library of engineering and computer science.

Includes bibliographical references (pages 169-188).

1. Introduction -- 1.1 Warehouse-scale computers -- 1.2 Cost efficiency at scale -- 1.3 Not just a collection of servers -- 1.4 One data center vs. several -- 1.5 Why WSCs might matter to you -- 1.6 Architectural overview of WSCs -- 1.6.1 Servers -- 1.6.2 Storage -- 1.6.3 Networking fabric -- 1.6.4 Buildings and infrastructure -- 1.6.5 Power usage -- 1.6.6 Handling failures and repairs -- 1.7 Overview of book --

2. Workloads and software infrastructure -- 2.1 Warehouse data center systems stack -- 2.2 Platform-level software -- 2.3 Cluster-level infrastructure software -- 2.3.1 Resource management -- 2.3.2 Cluster infrastructure -- 2.3.3 Application framework -- 2.4 Application-level software -- 2.4.1 Workload diversity -- 2.4.2 Web search -- 2.4.3 Video serving -- 2.4.4 Scholarly article similarity -- 2.4.5 Machine learning -- 2.5 Monitoring infrastructure -- 2.5.1 Service-level dashboards -- 2.5.2 Performance debugging tools -- 2.5.3 Platform-level health monitoring -- 2.6 WSC software tradeoffs -- 2.6.1 Data center vs. desktop -- 2.6.2 Performance and availability toolbox -- 2.6.3 Buy vs. build -- 2.6.4 Tail-tolerance -- 2.6.5 Latency numbers that engineers should know -- 2.7 Cloud computing -- 2.7.1 WSC for public cloud services vs. internal workloads -- 2.7.2 Cloud native software -- 2.8 Information security at warehouse scale --

3. WSC hardware building blocks -- 3.1 Server hardware -- 3.1.1 Server and rack overview -- 3.1.2 The impact of large SMP communication efficiency -- 3.1.3 Brawny vs. wimpy servers -- 3.2 Computing accelerators -- 3.2.1 GPUs -- 3.2.2 TPUs -- 3.3 Networking -- 3.3.1 Cluster networking -- 3.3.2 Host networking -- 3.4 Storage -- 3.4.1 Disk trays and diskless servers -- 3.4.2 Unstructured WSC storage -- 3.4.3 Structured WSC storage -- 3.4.4 Interplay of storage and networking technology -- 3.5 Balanced designs -- 3.5.1 System balance: storage hierarchy -- 3.5.2 Quantifying latency, bandwidth, and capacity --

4. Data center basics: building, power, and cooling -- 4.1 Data center overview -- 4.1.1 Tier classifications and specifications -- 4.1.2 Building basics -- 4.2 Data center power systems -- 4.2.1 Uninterruptible power systems (UPS) -- 4.2.2 Power distribution units (PDUs) -- 4.2.3 Comparison of AC and DC distribution architectures -- 4.3 Example: radial power distribution with redundancy -- 4.4 Example: medium voltage power plane -- 4.5 Data center cooling systems -- 4.5.1 Computer room air conditioners (CRACs) -- 4.5.2 Chillers -- 4.5.3 Cooling towers -- 4.5.4 Free cooling -- 4.5.5 Air flow considerations -- 4.5.6 In-rack, in-row, and liquid cooling -- 4.5.7 Container-based data centers -- 4.6 Example: Google's ceiling-mounted cooling for the data center -- 4.7 Summary --

5. Energy and power efficiency -- 5.1 Data center energy efficiency -- 5.1.1 The PUE metric -- 5.1.2 Issues with the PUE metric -- 5.1.3 Sources of efficiency losses in data centers -- 5.1.4 Improving the energy efficiency of data centers -- 5.1.5 Beyond the facility -- 5.2 The energy efficiency of computing -- 5.2.1 Measuring energy efficiency -- 5.2.2 Server energy efficiency -- 5.2.3 Usage profile of warehouse-scale computers -- 5.3 Energy-proportional computing -- 5.3.1 Causes of poor energy proportionality -- 5.3.2 Improving energy proportionality -- 5.3.3 Energy proportionality in the rest of the system -- 5.3.4 Relative effectiveness of low-power modes -- 5.3.5 The role of software in energy proportionality -- 5.4 Energy efficiency through specialization -- 5.5 Data center power provisioning -- 5.5.1 Deploying the right amount of equipment -- 5.5.2 Oversubscribing facility power -- 5.6 Trends in server energy usage -- 5.6.1 Using energy storage for power management -- 5.7 Summary --

6. Modeling costs -- 6.1 Capital costs -- 6.2 Operational costs -- 6.3 Case studies -- 6.4 Real-world data center costs -- 6.5 Modeling a partially filled data center -- 6.6 The cost of public clouds --

7. Dealing with failures and repairs -- 7.1 Implications of software fault tolerance -- 7.2 Categorizing faults -- 7.2.1 Fault severity -- 7.2.2 Causes of service-level faults -- 7.3 Machine-level failures -- 7.3.1 What causes machine crashes? -- 7.3.2 Predicting faults -- 7.4 Repairs -- 7.5 Tolerating faults, not hiding them -- 7.6 Accounting for faults in cluster system design --

8. Closing remarks -- 8.1 Hardware -- 8.2 Software -- 8.3 Economics and energy efficiency -- 8.4 Building responsive large-scale systems -- 8.4.1 Continually evolving workloads -- 8.4.2 Amdahl's cruel law -- 8.4.3 The attack of the killer microseconds -- 8.4.4 Tail at scale -- 8.5 Looking ahead -- 8.5.1 The ending of Moore's law -- 8.5.2 Accelerators and full system designs -- 8.5.3 Software-defined infrastructure -- 8.5.4 A new era of computer architecture and WSCs -- 8.6 Conclusions -- Bibliography -- Author biographies.

Abstract freely available; full-text restricted to subscribers or individual document purchasers.

Compendex

INSPEC

Google scholar

Google book search

This book describes warehouse-scale computers (WSCs), the computing platforms that power cloud computing and all the great web services we use every day. It discusses how these new systems treat the datacenter itself as one massive computer designed at warehouse scale, with hardware and software working in concert to deliver good levels of internet service performance. The book details the architecture of WSCs and covers the main factors influencing their design, operation, and cost structure, and the characteristics of their software base. Each chapter contains multiple real-world examples, including detailed case studies and previously unpublished details of the infrastructure used to power Google's online services. Targeted at the architects and programmers of today's WSCs, this book provides a great foundation for those looking to innovate in this fascinating and important area, but the material will also be broadly interesting to those who just want to understand the infrastructure powering the internet. The third edition reflects four years of advancements since the previous edition and nearly doubles the number of pictures and figures. New topics range from additional workloads like video streaming, machine learning, and public cloud to specialized silicon accelerators, storage and network building blocks, and a revised discussion of data center power and cooling, and uptime. Further discussions of emerging trends and opportunities ensure that this revised edition will remain an essential resource for educators and professionals working on the next generation of WSCs.

Also available in print.

Title from PDF title page (viewed on November 1, 2018).