IT 344 Operating Systems Distributed Systems What

IT 344: Operating Systems Distributed Systems

What is a “distributed system”? • Very broad definition – loosely-coupled to tightly-coupled • Nearly all systems today are distributed in some way – – – – they use email they access files over a network they access printers over a network they’re backed up over a network they share other physical or logical resources they cooperate with other people on other machines they access the web they receive video, audio, etc. 3/16/2018 2

Distributed systems are now a requirement • Economics dictate that we buy small computers • Everyone needs to communicate • We need to share physical devices (printers) as well as information (files, etc. ) • Many applications are by their nature distributed (bank teller machines, airline reservations, ticket purchasing) • To solve the largest problems, we will need to get large collections of small machines to cooperate together (parallel programming) 3/16/2018 3

Loosely-coupled systems • Earliest systems used simple explicit network programs – FTP (rcp): file transfer program – telnet (rlogin/rsh): remote login program – mail (SMTP) • Each system was a completely autonomous independent system, connected to others on the network 3/16/2018 4

• Even today, most distributed systems are looselycoupled – – – each CPU runs an independent autonomous OS computers don’t really trust each other some resources are shared, but most are not the system may look differently from different hosts typically, communication times are long 3/16/2018 5

Closely-coupled systems • A distributed system becomes more “closely-coupled” as it – – – appears more uniform in nature runs a “single” operating system has a single security domain shares all logical resources (e. g. , files) shares all physical resources (CPUs, memory, disks, printers, etc. ) • In the limit, a distributed system looks to the user as if it were a centralized timesharing system, except that it’s constructed out of a distributed collection of hardware and software components 3/16/2018 6

Tightly-coupled systems • A “tightly-coupled” system usually refers to a multiprocessor – runs a single copy of the OS with a single job queue – has a single address space – usually has a single bus or backplane to which all processors and memories are connected – has very low communication latency – processors communicate through shared memory 3/16/2018 7

Some issues in distributed systems • • Transparency (how visible is the distribution) Security Reliability Performance Scalability Programming models Communication models 3/16/2018 8

Distributed File Systems • The most common distributed services: – – printing email Files Computation • Basic idea of distributed file systems – support network-wide sharing of files and devices (disks) • Generally provide a “traditional” view – a centralized shared local file system • But with a distributed implementation – read blocks from remote hosts, instead of from local disks 3/16/2018 9

Issues • What is the basic abstraction – remote file system? • open, close, read, write, … – remote disk? • read block, write block • Naming – how are files named? – are those names location transparent? • is the file location visible to the user? – are those names location independent? • do the names change if the file moves? • do the names change if the user moves? 3/16/2018 10

• Caching – caching exists for performance reasons – where are file blocks cached? • on the file server? • on the client machine? • both? • Sharing and coherency – what are the semantics of sharing? – what happens when a cached block/file is modified – how does a node know when its cached blocks are out of date? 3/16/2018 11

• Replication – – replication can exist for performance and/or availability can there be multiple copies of a file in the network? if multiple copies, how are updates handled? what if there’s a network partition and clients work on separate copies? • Performance – – what is the cost of remote operation? what is the cost of file sharing? how does the system scale as the number of clients grows? what are the performance limitations: network, CPU, disks, protocols, data copying? 3/16/2018 12

Example: SUN Network File System (NFS) • The Sun Network File System (NFS) has become a common standard for distributed UNIX file access • NFS runs over LANs (even over WANs – slowly) • Basic idea – allow a remote directory to be “mounted” (spliced) onto a local directory – Gives access to that remote directory and all its descendants as if they were part of the local hierarchy • Pretty much exactly like a “local mount” or “link” on UNIX – except for implementation and performance … – no, we didn’t really learn about these, but they’re obvious 3/16/2018 13

• For instance: – I mount /u 4/teng on Node 1 onto /students/foo on Node 2 – users on Node 2 can then access this directory as /students/foo – if I had a file /u 4/teng/myfile, users on Node 2 see it as /students/foo/myfile • Just as, on a local system, I might link /groups/it 344/www/10 wi/ as /u 4/teng/it 344 to allow easy access to my web data from class home directory 3/16/2018 14

NFS implementation • NFS defines a set of RPC operations for remote file access: – – searching a directory reading directory entries manipulating links and directories reading/writing files • Every node may be both a client and server 3/16/2018 15

• NFS defines new layers in the Unix file system System Call Interface The virtual file system (VFS) provides a standard interface, using v-nodes as file handles. A v-node describes either a local or remote file. Virtual File System UFS (local files) NFS (remote files) RPCs to other (server) nodes RPC requests from remote clients, and server responses buffer cache / i-node table 3/16/2018 16

NFS caching / sharing • On an open, the client asks the server whether its cached blocks are up to date. • Once a file is open, different clients can write it and get inconsistent data. • Modified data is flushed back to the server every 30 seconds. 3/16/2018 17

Example: CMU’s Andrew File System (AFS) • Developed at CMU to support all of its student computing • Consists of workstation clients and dedicated file server machines (differs from NFS) • Workstations have local disks, used to cache files being used locally (originally whole files, subsequently 64 K file chunks) (differs from NFS) • Andrew has a single name space – your files have the same names everywhere in the world (differs from NFS) • Andrew is good for distant operation because of its local disk caching: after a slow startup, most accesses are to local disk 3/16/2018 18

AFS caching/sharing • Need for scaling required reduction of client-server message traffic • Once a file is cached, all operations are performed locally • On close, if the file has been modified, it is replaced on the server • The client assumes that its cache is up to date, unless it receives a callback message from the server saying otherwise – on file open, if the client has received a callback on the file, it must fetch a new copy; otherwise it uses its locally-cached copy (differs from NFS) 3/16/2018 19

Example: Berkeley Sprite File System • Unix file system developed for diskless workstations with large memories at UCB (differs from NFS, AFS) • Considers memory as a huge cache of disk blocks – memory is shared between file system and VM • Files are permanently stored on servers – servers have a large memory that acts as a cache as well • Several workstations can cache blocks for read-only files • If a file is being written by more than 1 machine, client caching is turned off – all requests go to the server (differs from NFS, AFS) 3/16/2018 20

Other Approaches • Serverless – Xfs, Farsite • Highly Available – GFS • Mostly Read Only – WWW • State, not Files – SQL Server – Big. Table 3/16/2018 21

Administrivia • Case study http: //www. et. byu. edu/groups/it 344/10 wi/casestudies. ht m#_Assignments • Lab 8 – last one of the semester, no write up, pass off gives you full credit • No lab next week – catch up on past due work and work on BYOOS • BYOOS part 3 – write up just to show progress • HW 10 – last one of the semester • Final exam, on blackboard, date TBD, probably 1 st week of April • HONOR CODE 3/16/2018 22

Example: Google File System (GFS) Independence Small Scale Many users Many programs 3/16/2018 Cooperation

“Google” circa 1997 (google. stanford. edu) 3/16/2018 24

Google (circa 1999) 3/16/2018 25

Google data center (circa 2000) 3/16/2018 26

Google new data center 2001 3/16/2018 27

Google data center (3 days later) 3/16/2018 28

GFS: Google File System • Why did Google build its own FS? • Google has unique FS requirements – – Huge read/write bandwidth Reliability over thousands of nodes Mostly operating on large data blocks Need efficient distributed operations • Unfair advantage – Google has control over applications, libraries and operating system 3/16/2018 29

GFS Idealogy • Huge amount of data • Ability to efficiently access data w/ low locality, typical query reads 100 s MB of data • Large quantity of Cheap machines: performance vs performance/$, performance/W • Replication: scalability and h/w failure • BW more important than latency • Component failures are the norm rather than the exception • Atomic append operation so that multiple clients can append concurrently 3/16/2018 30

GFS Usage @ Google • • • 200+ clusters Filesystem clusters of 1000 s of machines Pools of 1000+ clients 4+ PB Filesystems 40 GB/s read/write load (in the presence of frequent HW failures) 3/16/2018 32

Files in GFS • Files are huge by traditional standards • Most files are mutated by appending new data rather than overwriting existing data • Once written, the files are only read, and often only sequentially. • Appending becomes the focus of performance optimization and atomicity guarantees 3/16/2018 33

Masters C 0 C 5 C 1 C 2 Chunkserver 1 • • • Replicas GFS Setup Client GFS Master C 1 C 5 Misc. servers GFS Master C 3 Chunkserver 2 Client C 0 … C 5 C 2 Chunkserver N Master manages metadata Data transfers happen directly between clients/chunkservers Files broken into chunks (typically 64 MB) 3/16/2018 34

Architecture • GFS cluster consists of a single master and multiple chunk servers and is accessed by multiple clients. • Each of these is typically a commodity Linux machine running a user-level server process. • Files are divided into fixed-size chunks identified by an immutable and globally unique 64 bit chunk handle • For reliability, each chunk is replicated on multiple chunk servers • master maintains all file system metadata. • The master periodically communicates with each chunk server in Heart. Beat (timer) messages to give it instructions and collect its state • Neither the client nor the chunk server caches file data eliminating cache coherence issues. • Clients do cache metadata, however. 3/16/2018 35

Architecture 3/16/2018 36

Read Process • Single master vastly simplifies design • Clients never read and write file data through the master. Instead, a client asks the master which chunk servers it should contact. • Using the fixed chunk size, the client translates the file name and byte offset specified by the application into a chunk index within the file • It sends the master a request containing the file name and chunk index. The master replies with the corresponding chunk handle and locations of the replicas. The client caches this information using the file name and chunk index as the key. • The client then sends a request to one of the replicas, most likely the closest one. The request specifies the chunk handle and a byte range within that chunk 3/16/2018 37

Specifications • Chunk Size = 64 MB • Chunks stored as plain Unix files on chunk server. • A persistent TCP connection to the chunk server over an extended period of time (reduce network overhead) • cache all the chunk location information to facilitate small random reads. • Master keeps the metadata in memory • Disadvantages – Small files become Hotspots. • Solution – Higher replication for such files. 3/16/2018 38

Microsoft Data Center 4. 0 • http: //www. youtube. com/watch? v=PPno. Kb 9 f. Tk. A 3/16/2018 39

Data center container • Microsoft $500 M Chicago data center (2009) • > 2000 servers/container (40 ft) • 150 containers • 11 diesel generators, each 2. 8 megawatts • 12 chillers, each 1260 tons 3/16/2018 40

Data center container • • Google IBM HP … 3/16/2018 41

3/16/2018 42

Cloud Computing Platforms 3/16/2018 43

Client/server computing • • Mail server/service File server/service Print server/service Compute server/service Game server/service Music server/service Web server/service etc. 3/16/2018 44

Peer-to-peer (p 2 p) systems • Napster • Gnutella (Lime. Wire) – example technical challenge: selforganizing overlay network – technical advantage of Gnutella? – er … legal advantage of Gnutella? Data source: Digital Music News Research Group 3/16/2018 45

Summary • There a number of issues to deal with: – – – what is the basic abstraction naming caching sharing and coherency replication performance • No right answer! Different systems make different tradeoffs! 3/16/2018 46

• Performance is always an issue – always a tradeoff between performance and the semantics of file operations (e. g. , for shared files). • Caching of file blocks is crucial in any file system – maintaining coherency is a crucial design issue. • Newer systems are dealing with issues such as disconnected operation for mobile computers 3/16/2018 47

Service Oriented Architecture • How do you allow hundreds of developers to work on a single website?

Amazon. com: The Beginning • Initially, one web server (Obidos) and one database Internet Obidos Database l Details: Front end consists of a web server (Apache) and “business logic” (Obidos)

Amazon: Success Disaster! Use redundancy to scale-up, improve availability Obidos Amazon. com Internet Load balancer Obidos Database Obidos

Obidos • Obidos was a single monolithic C application that comprised most of Amazon. com’s functionality • During scale-up, this model began to break down

Problem #1: Branch Management • Merging code across branches becomes untenable Hello. World. c release development Blue changes depend on Red changes (which may depend on other changes…)

Problem #2: Debugging • On a failure, we would like to inspect what happened “recently” – But, the change log contains numerous updates from many groups • Bigger problem: lack of isolation – Change by one group can impact others

Problem #3: Linker Failure • Obidos grew so large that standard build tools were failing

Service-Oriented Architecture (1) • First, decompose the monolithic web site into a set of smaller modules – Called services • Examples: – – Recommendation service Price service Catalogue service And MANY others

Sidebar: Modularity • Information hiding (Parnas 1972): The purpose of a module is to hide secrets public interface List { } // This can be an array, a linked-list, // or something else • Benefits of modularity – Groups can work independently • Less “synchronization overhead” – Ease of change • We are free to change the hidden secrets – Ease of comprehension • Can study the system at a high level of abstraction

Systems and Information Hiding • There is often a tension between performance and information hiding • In OS’s, performance often wins: struct buffer { // DO NOT MOVE these fields! // They are accessed by inline assembly that // assumes the current ordering. struct buffer* next; struct buffer* prev; int size; … }

Service Oriented Architectures (2) • Modularity + a network • Services live on disjoint sets of machines • Services communicate using RPC – Remote procedure call

Remote Procedure Call • RPC exposes a programming interface across machines: interface Price. Service { float get. Price(long unique. ID); } Price. Impl Server Client get. Price()

SOA, Visualized Shopping Cart Price Website Recommendation • All services reside on separate machines • All invocations are remote procedure calls Catalogue

Benefits of SOA • Modularity and service isolation – This extends all the way down to the OS, programming language, build tools, etc. • Better visibility – Administrators can monitor the interactions between services • Better resource accounting – Who is using which resources?

Performance Issues • A webpage can require dozens of service calls – RPC system must be high performance • Metrics of interest: – Throughput – Latency • Both average and the variance

SLAs • Service performance is dictated by contracts called Service Level Agreements – e. g. , Service Foo must • Have 4 9’s of availability • Have a median latency of 50 ms • Have a 3 9’s latency of 200 ms

Amazon and Web Services Sleds. com Front-end website Catalogue Amazon. com Order Processing Shopping Carts • Allow third-parties to use some (but not all) of the Amazon platform

Searching on a Web Site

Searching Through a Web Service class Program { static void Main(string[] args) { AWSECommerce. Service service = new AWSECommerce. Service(); Item. Search request = new Item. Search(); request. Subscription. Id = "0525 E 2 PQ 81 DD 7 ZTWTK 82"; request. Request = new Item. Search. Request[1]; request. Request[0] = new Item. Search. Request(); request. Request[0]. Response. Group = new string[] { "Small" }; request. Request[0]. Search. Index = "Books"; request. Request[0]. Author = "Tom Clancy"; Item. Search. Response response = service. Item. Search(request); Console. Write. Line(response. Items[0]. Item. Length + " books written by Tom Clancy found. "); } }

Other Web Services • Google – Calendar – Maps – Charts • Amazon infrastructure services (cloud) – Simple storage (disk) – Elastic compute cloud (virtual machines) – Simple. DB • Facebook • Ebay • …