The Datacenter as a Computer (Chapters 1, 2)

Motivation:

Introduce the concept of warehouse scale computers and outline the

differences with traditional datacenters. The first two chapters

introduce the architecture and software components such clusters.

Main Ideas:

- There is a broad shift to server-side computation due to the ease of

deployment and the growing influence of large scale workloads like

websearch.

- Warehouse-scale computers are defined as clusters consisting of

thousands of machines and services which are deployed across them.

With such large scale, failures are commonplace and are a part of the

design.

- The architecture of the warehouse scale computer consists of a set of

servers arranged in racks and connected by a rack-level switch. The

racks are then connected by a cluster-level switch which could be

over-subscribed by a factor of 5 or 10.

- The bandwidth and latency varies a lot for memory / disk accesses

on a local machine vs. machines in the same rack vs. machines in other

racks.

- There are many levels of software components in a warehouse scale

computer. The resource manager, storage servers and monitoring

infrastructure are examples of lower-level software components

- Applications in warehouse scale clusters could be latency senstive

like websearch or batch processing jobs like indexing and article

similarity.

Trade-offs, influence

- This work analyzes many trade-offs involved in constructing a

warehouse scale computer. These include

- Using NAS storage solutions vs. distributed file system like GFS.

GFS provides better availability and performance benefits for locality

aware applications.

- Choice between buying and building software components for warehouse

scale computing. Building components in-house gives more flexibility

and ease of maintenance.

- Software components can also trade-off between advanced features and

performance. For example, Bigtable does not support multi-row

transactions and GFS does not have POSIX semantics. These are not

frequently used in the workloads observed at Google, Microsoft etc.

- The advent of warehouse scale computers has been a key driving factor

for the rise of cloud computing. With increasing operating costs for power

consumption, warehouse scale computers can help in building more

efficient systems to support newer applications.

However, there will be a client-side component for most of the

applications (the authors acknowledge this) in say a mobile device or

a web-browser. With changing hardware costs, there could be a shift in

the most efficient way to split the computation between the server and clients.

Above the Clouds: A Berkeley View of Cloud Computing

Problem Motivation:

- To define cloud computing and to analyze the trade-offs for cloud

providers and users

Main Ideas:

- Cloud computing refers to both applications delivered over the

Internet and hardware/software systems in datacenters.

- Computing available as a utility is an important shift in IT

infrastructure and similar to semiconductor foundries, it can enable

many more users to build scalable web-services.

- Cloud providers benefit from statistical multiplexing and bulk

purchasing. They can make money as it is cheaper to build larger datacenters and

cost is amortized. It also helps fulfill the needs of their existing customers

e.g., Microsoft, IBM etc.

- Advantage for users is that cloud computing allows them to scale their

applications based on traffic patterns. This includes scaling up and

scaling down and based on the interface, this could be automatic or

specified by the programmer.

- Economic factors have influenced the growth of cloud computing. Jim Gray

observed in 2003 that cost of wide area networking has fallen

more slowly compared to CPU and storage costs. Hence it is better to put

applications closer to the data. This trend continues (as seen by 2008

numbers) and when opeartional costs are included, it is cheaper to

rent than to buy clusters.

- There are ten obstacles identified for the adoption and growth of

cloud computing. Some of these are policy issues like licensing, while

others are research opportunities like a scalable storage solution.

Trade-offs, Influence

- The paper analyzes the economic trade-offs for an organization to use

cloud computing against building their own datacenter. This trade-off

is represented as an equation and based on the cluster utilization,

companies can decide when it would be profitable for them to switch to

the cloud.

- The cloud computing interface also represents a trade-off between

flexibility and ease of use. While Amazon EC2 instances, close to

bare-metal hardware, provide flexibility, it is not easy to get

automatic scale up / scale down.

On the other hand Google AppEngine is designed for a restricted set of

applications, but provides automatic scaling.

- With the growing adoption of cloud computing, the ideas presented in

this paper are influential today and I think they will also be influential

in the future.