Definition of a Distributed System 1 A distributed

Definition of a Distributed System (1) A distributed system is: A collection of independent computers that appears to its users as a single coherent system.

Definition of a Distributed System (2) users ? 1. 1 Standard OS A distributed system organized as middleware. Note that the middleware layer extends over multiple machines.

Goals • • • Access and share remote resources Interoperability Portability Flexibility Transparency Scalability

Transparency in a Distributed System Transparency Description Access Hide differences in data representation and how a resource is accessed Location Hide where a resource is located Migration Hide that a resource may move to another location Relocation Hide that a resource may be moved to another location while in use Replication Hide that a resource may be replicated Concurrency Hide that a resource may be shared by several competitive users Failure Hide the failure and recovery of a resource Persistence Hide whether a (software) resource is in memory or on disk Different forms of transparency in a distributed system.

Scalability • Size • Geography • Administrative organizations Scalability problems in a distributed system Concept Example Centralized services A single server for all users Centralized data A single on-line telephone book Centralized algorithms Doing routing based on complete information Decentralized algorithms characteristics • None has complete information about the system state • Machines take decisions on local info • Failure of one machine doesn’t affect the algorithm • There is no assumption about a global clock

Scalability How to solve scalability problems? a) Hiding communication latencies • Asynchronous communications (but not only, not always) b) Distribution c) Replication (with care for consistency)

Scaling Techniques (1) 1. 4 Moving part of the computation to the client…(hiding comm. latency) The difference between letting: a) a server or b) a client c) check forms as they are being filled

Scaling Techniques (2) 1. 5 Distribution… Replication… DNS name space divided into zones, with possibility of slave server

Software Concepts Operating systems for distributed computers System Description Main Goal DOS Tightly-coupled operating system for multiprocessors and homogeneous multicomputers Hide and manage hardware resources NOS Loosely-coupled operating system for heterogeneous multicomputers (LAN and WAN) Offer local services to remote clients Middleware Additional layer atop of NOS implementing general-purpose services Provide distribution transparency An overview between • DOS (Distributed Operating Systems) • NOS (Network Operating Systems) • Middleware

Uniprocessor Operating Systems 1. 11 Separating applications from operating system code through a microkernel.

Multicomputer Operating Systems (1) 1. 14 General structure of a multicomputer operating system Communication Message passing

Multicomputer Operating Systems (2) 1. 15 Alternatives for blocking and buffering in message passing • Buffering: sender and receiver side • 4 possible synchronization points Reliable communication?

Multicomputer Operating Systems (3) Synchronization point Send buffer Reliable comm. guaranteed? Block sender until buffer not full (S 1) Yes Not necessary Block sender until message sent (S 2) No Not necessary Block sender until message received (S 3) No Necessary Block sender until message delivered (S 4) No Necessary Relation between blocking, buffering, and reliable communications.

Network Operating System (1) 1 -19 General structure of a network (not distributed) operating system. Users see the structure

Network Operating System (2) Evolution : file server 1 -20 Two clients and a file server in a network operating system. File servers generally maintain hierarchical file system.

Network Operating System (3) 1. 21 Different clients may mount the servers in different places. Different clients may have different views of the file systems

With a basic network operating system we loose in transparency • Explicit remote login • Independent accounting • Independent access permission • Low protection

Positioning Middleware 1 -22 General structure of a distributed system as middleware. Middleware does not manage an individual node

Middleware Models • Plan 9 – file oriented paradigm (Unix like) • Distributed file system – transparency supported only for traditional files • Models based on RPCs – hide network communications • Distributed objects – each object can be located on a single machine, each object interface hides internal details including communications • Distributed documents - WWW

Middleware Services • Naming allows entities to be shared and looked up. • Persistence for storage, i. e. databases or facilities to connect to databases • Distributed transactions allow multiple read and write operations to occur atomically. • Security • Access transparency by communication facilities hiding message passing through network

Middleware and Openness Interfaces definition has to be complete. Incompleteness lead to bad portability and bad interoperability 1. 23 In an open middleware-based distributed system, the protocols used by each middleware layer should be the same, as well as the interfaces they offer to applications.

Comparison between Systems Distributed OS Item Network OS Middlewarebased OS Multiproc. Multicomp. Very High Low High Yes No No Number of copies of OS 1 N N N Basis for communication Shared memory Messages/Fi les Model specific Resource management Global, central Global, distributed Per node Scalability No Moderately Yes Varies Openness Closed Open… Degree of transparency Same OS on all nodes In green open problems are emphasized

Clients and Servers Client-server model is widely used to understand the complexity of distributed systems processes 1. 25 General interaction between a client and a server. Request-reply behavior What kind of protocol for communication?

Clients and Servers • Connectionless protocol – Efficient – Not reliable • Retransmission on “failure” can be dangerous • Connection oriented protocol – Lower performance – reliable

The three Levels for client server model How can we make a distinction between client and server ? 1 -28 The general organization of an Internet search engine into three different layers (we can have this organization on two or more physically distributed machines)

Multitiered Architectures (1) 1 -29 Alternative client-server organizations (two-tiered architecture)

Multitiered Architectures (2) 1 -30 An example of a server acting as a client( vertical distribution). Three-tiered architecture

Modern Architectures Often, clients and servers distribution counts more then different logical components Request from client 1 -31 An example of horizontal distribution of a Web service.