chapter 16 distributed processing, client/server, and clusters

structured query language (SQL)

A language developed by IBM and standardized by ANSI for addressing, creating, updating, or querying relational databases.

relational database

A database in which information access is limited to the selection of rows that satisfy all search criteria.

cache consistency problem

. The problem of keeping local cache copies up to date to changes in remote data The simplest approach to cache consistency is to use file-locking techniques to prevent simultaneous access to a file by more than one client. This guarantees consistency at the expense of performance and flexibility. A more powerful approach is provided with the facility in Sprite [NELS88, OUST88]. Any number of remote processes may open a file for read and create their own client cache. But when an open file request to a server requests write access and other processes have the file open for read access, the server takes two actions. First, it notifies the writing process that, although it may maintain a cache, it must write back all altered blocks immediately upon update. There can be at most one such client. Second, the server notifies all reading processes that have the file open that the file is no longer cacheable.

server

A computer, usually a high-powered workstation, a minicomputer, or a mainframe, that houses information for manipulation by networked clients. Each server in the client/server environment provides a set of shared services to the clients. The most common type of server currently is the database server, usually controlling a relational database. The server enables many clients to share access to the same database and enables the use of a high-performance computer system to manage the database.

client

A networked information requester, usually a PC or workstation, that can query database and/or other information from a serve The client machines are generally single-user PCs or workstations that provide a highly user-friendly interface to the end user. The client-based station generally presents the type of graphical interface that is most comfortable to users, including the use of windows and a mouse. Microsoft Windows and Macintosh OS provide examples of such interfaces. Client-based applications are tailored for ease of use and include such familiar tools as the spreadsheet.

failback

A related function is the restoration of applications and data resources to the original system once it has been fixed; this is referred to as failback. Failback can be automated, but this is desirable only if the problem is truly fixed and unlikely to recur. If not, automatic failback can cause subsequently failed resources to bounce back and forth between computers, resulting in performance and recovery problems.

middleware

A set of drivers, APIs, or other software that improves connectivity between a client application and a server

applications programming interface (API)

A set of function and call programs that allow clients and servers to intercommunicate.

remote procedure call (RPC)

A variation on the basic message-passing model is the remote procedure call. This is now a widely accepted and common method for encapsulating communication in a distributed system. The essence of the technique is to allow programs on different machines to interact using simple procedure call/return semantics, just as if the two programs were on the same machine. That is, the procedure call is used for access to remote services. The popularity of this approach is due to the following advantages. 1. The procedure call is a widely accepted, used, and understood abstraction. 2. The use of remote procedure calls enables remote interfaces to be specified as a set of named operations with designated types. Thus, the interface can be clearly documented and distributed programs can be statically checked for type errors. 3. Because a standardized and precisely defined interface is specified, the communication code for an application can be generated automatically. 4. Because a standardized and precisely defined interface is specified, developers can write client and server modules that can be moved among computers and operating systems with little modification and recoding. The remote procedure call mechanism can be viewed as a refinement of reliable, blocking message passing

GUI (graphical user interface)

An essential factor in the success of a client/server environment is the way in which the user interacts with the system as a whole. Thus, the design of the user interface on the client machine is critical. In most client/server systems, there is heavy emphasis on providing a graphical user interface (GUI) that is easy to use, easy to learn, yet powerful and flexible. Thus, we can think of a presentation services module in the client workstation that is responsible for providing a user-friendly interface to the distributed applications available in the environment.

client based processing

At the other extreme, virtually all application processing may be done at the client, with the exception of data validation routines and other database logic functions that are best performed at the server. Generally, some of the more sophisticated database logic functions are housed on the client side. This architecture is perhaps the most common client/server approach in current use. It enables the user to employ applications tailored to local needs.

cluster

Clustering is an alternative to symmetric multiprocessing (SMP) as an approach to providing high performance and high availability and is particularly attractive for server applications. We can define a cluster as a group of interconnected, whole computers working together as a unified computing resource that can create the illusion of being one machine. The term whole computer means a system that can run on its own, apart from the cluster; in the literature, each computer in a cluster is typically referred to as a node. [BREW97] lists four benefits that can be achieved with clustering. These can also be thought of as objectives or design requirements: Absolute scalability: It is possible to create large clusters that far surpass the power of even the largest stand-alone machines. A cluster can have dozens or even hundreds of machines, each of which is a multiprocessor. Incremental scalability: A cluster is configured in such a way that it is possible to add new systems to the cluster in small increments. Thus, a user can start out with a modest system and expand it as needs grow, without having to go through a major upgrade in which an existing small system is replaced with a larger system. High availability: Because each node in a cluster is a stand-alone computer, the failure of one node does not mean loss of service. In many products, fault tolerance is handled automatically in software. Superior price/performance: By using commodity building blocks, it is possible to put together a cluster with equal or greater computing power than a single large machine, at much lower cost.

host based approach

Host-based processing is not true client/server computing as the term is generally used. Rather, host-based processing refers to the traditional mainframe environment in which all or virtually all of the processing is done on a central host. Often the user interface is via a dumb terminal. Even if the user is employing a microcomputer, the user's station is generally limited to the role of a terminal emulator.

beowulf

In 1994, the Beowulf project was initiated under the sponsorship of the NASA High Performance Computing and Communications (HPCC) project. Its goal was to investigate the potential of clustered PCs for performing important computation tasks beyond the capabilities of contemporary workstations at minimum cost. Today, the Beowulf approach is widely implemented and is perhaps the most important cluster technology available. Beowulf Features Key features of Beowulf include the following [RIDG97]: Mass market commodity components Dedicated processors (rather than scavenging cycles from idle workstations) A dedicated, private network (LAN or WAN or internetted combination) \ No custom components Easy replication from multiple vendors Scalable I/O A freely available software base Use of freely available distribution computing tools with minimal changes Return of the design and improvements to the community

cooperative processing

In a cooperative processing configuration, the application processing is performed in an optimized fashion, taking advantage of the strengths of both client and server machines and of the distribution of data. Such a configuration is more complex to set up and maintain but, in the long run, this type of configuration may offer greater user productivity gains and greater network efficiency than other client/server approaches.

distributed message passing

It is usually the case in a distributed processing systems that the computers do not share main memory; each is an isolated computer system. Thus, interprocessor communication techniques that rely on shared memory, such as semaphores, cannot be used. Instead, techniques that rely on message passing are used. In this section and the next, we look at the two most common approaches. The first is the straightforward application of messages as they are used in a single system. The second is a separate technique that relies on message passing as a basic function: the remote procedure call. Figure 16.10a shows the use of message passing to implement client/server functionality. A client process requires some service (e.g., read a file, print) and sends a message containing a request for service to a server process. The server process honors the request and sends a message containing a reply. In its simplest form, only two functions are needed: Send and Receive. The Send function specifies a destination and includes the message content. The Receive function tells from whom a message is desired (including "all") and provides a buffer where the incoming message is to be stored.

failover

The function of switching an application and data resources over from a failed system to an alternative system in the cluster is referred to as failover

server based approach

The most basic class of client/server configuration is one in which the client is principally responsible for providing a graphical user interface, while virtually all of the processing is done on the server. This configuration is typical of early client/server efforts, especially departmental-level systems. The rationale behind such configurations is that the user workstation is best suited to providing a user-friendly interface and that databases and applications can easily be maintained on central systems. Although the user gains the advantage of a better interface, this type of configuration does not generally lend itself to any significant gains in productivity or to any fundamental changes in the actual business functions that the system supports.

thin client approach

This approach more nearly mimics the traditional host-centered approach and is often the migration path for evolving corporate-wide applications from the mainframe to a distributed environment.

file cache consistency

When caches always contain exact copies of remote data, we say that the caches are consistent. It is possible for caches to become inconsistent when the remote data are changed and the corresponding obsolete local cache copies are not discarded.

fat client

a considerable fraction of the load is on the client. This so-called fat client model has been popularized by application development tools such as Sybase Inc.'s PowerBuilder and Gupta Corp.'s SQL Windows. Applications developed with these tools are typically departmental in scope. The main benefit of the fat client model is that it takes advantage of desktop power, offloading application processing from servers and making them more efficient and less likely to be bottlenecks. There are, however, several disadvantages to the fat client strategy. The addition of more functions rapidly overloads the capacity of desktop machines, forcing companies to upgrade. If the model extends beyond the department to incorporate many users, the company must install high-capacity LANs to support the large volumes of transmission between the thin servers and the fat clients. Finally, it is difficult to maintain, upgrade, or replace applications distributed across tens or hundreds of desktops.