Transcript Multiprocessor and Distributed Systems

Multiprocessing and Distributed Systems
CS-502 Operating Systems
Fall 2006
(Slides include materials from Operating System Concepts, 7th ed., by Silbershatz, Galvin, & Gagne,
Modern Operating Systems, 2nd ed., by Tanenbaum, and
Operating Systems: Internals and Design Principles, 5th ed., by Stallings)
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
1
Multiprocessing  Distributed Computing
(a spectrum)
• Many independent problems at same time
• Similar
• Different
• One very big problem (or a few)
• Computations that are physically separated
• Client-server
• Inherently dispersed computations
Different kinds of computers and operating systems
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
2
Multiprocessing  Distributed Computing
(a spectrum)
• Many independent problems at same time
• Similar — e.g., banking & credit card; airline reservations
• Different — e.g., university computer center; your own PC
• One very big problem (or a few)
• Computations that are physically separated
• Client-server
• Inherently dispersed computations
Different kinds of computers and operating systems
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
3
Multiprocessing  Distributed Computing
(a spectrum)
• Many independent problems at same time
• Similar — e.g., banking & credit card; airline reservations
• Different — e.g., university computer center; your own PC
• One very big problem (too big for one computer)
• Weather modeling, finite element analysis; drug discovery;
gene modeling; weapons simulation; etc.
• Computations that are physically separated
• Client-server
• Inherently dispersed computations
Different kinds of computers and operating systems
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
4
Multiprocessing  Distributed Computing
(a spectrum)
• Many independent problems at same time
• Similar — e.g., banking & credit card; airline reservations
• Different — e.g., university computer center; your own PC
• One very big problem (too big for one computer)
• Weather modeling, Finite element analysis; Drug discovery;
Gene modeling; Weapons simulation; etc.
• Computations that are physically separated
• Client-server
• Dispersed – routing tables for internet; electric power distrib.
Different kinds of computers and operating systems
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
5
Taxonomy of Parallel Processing
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
6
Inherently Distributed Computation
(example)
• Internet routing of packets
• Network layer – send a packet to its IP address
• Problems
• No central authority to manage routing in Internet
• Nodes and links come online or fail rapidly
• Need to be able to send a packet from any address to
any address
• Solution
• Distributed routing algorithm
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
7
Distributed routing algorithm
(simplified example)
• Each node “knows” networks other nodes
directly connected to it.
• Each node maintains table of distant
networks
• [network #, 1st hop, distance]
• Periodically adjacent nodes exchange tables
• Update algorithm (for each network in table)
• If (neighbor’s distance to network < my distance to
network + my distance to neighbor), then update my
table entry for that network
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
8
Distributed routing algorithm
(result)
• All nodes in Internet have reasonably up-todate routing tables
• Rapid responses to changes in network
topology, congestion, failures, etc.
• Very reliable with no central management!
• Network management software
• Monitoring health of network (e.g., routing tables)
• Identifying actual or incipient problems
• Data and statistics for planning purposes
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
9
Summary
• Some algorithms are inherently distributed
• Depend upon computations at physically
separate places
• Result is the cumulative results of
individual computations
• Not much impact on computer or OS
architecture
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
10
Client-Server Computations
• Very familiar model
• Parts of computation takes place on
• User’s PC, credit-card terminal, etc.
• … and part takes place on central server
• E.g., updating database, search engine, etc.
• Central issues
• Protocols for efficient communication
• Where to partition the problem
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
11
Very Big Problems
• Lots and lots of calculations
• Highly repetitive
• Easily parallelizable into separate subparts
• (Usually) dependent upon adjacent subparts
• Arrays or clusters of computers
• Independent memories
• Very fast communication between elements
 close coupling
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
12
Closely-coupled Systems
• Examples
– Beowulf clusters – 32-256 PCs
– ASCI Red – ~1000-4000 PC-like processors
– BlueGene/L – >16,000 processing elements
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
13
Taxonomy of Parallel Processing
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
14
IBM BlueGene/L
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
15
Systems for Closely-Coupled Clusters
• Independent OS for each node
• E.g., Linux in Beowulf
• Additional middleware for
•
•
•
•
Distributed process space
Synchronization among nodes (Silbershatz, Ch 18)
Access to network and peripheral devices
Failure management
• Parallelizing compilers
• For partitioning a problem into many processes
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
16
Cluster Computer Architecture
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
17
Interconnection Topologies
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
18
Goal
• Microsecond-level
– messages
– synchronization and barrier operations among
processes
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
19
Questions?
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
20
Multiprocessing  Distributed Computing
(a spectrum)
• Many independent problems at same time
• Similar — e.g., banking & credit card; airline reservations
• Different — e.g., university computer center; your own PC
• One very big problem (or a few)
• Problems that physically separated by their very
nature
Different kinds of computers and operating systems
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
21
Many Separate Computations
• Similar – banking & credit card; airline reservation; …
• Transaction processing
• Thousands of small transactions per second
• Few (if any) communications among transactions
– Exception: locking, mutual exclusion
• Common database
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
22
Requirements
• Lots of disks, enough CPU power
• Accessible via multiple paths
• High availability
• Fault-tolerant components & OS support
• Processor independence
•
•
•
•
CS-502 Fall 2006
Any transaction on any processor
Global file system
Load-balancing
Scalable
Multiprocessing and
Distributed Systems
23
Options
• Cluster (of a somewhat different sort)
• Shared-memory multi-processor
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
24
Taxonomy of Parallel Processing
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
25
Cluster Computing
• Scalable to large number of processors
• Common I/O space
– Multiple channels for disk access
– Hyper-channel, Fiber-channel, etc.
• Multi-headed disks
– At least two paths from any processor to any disk
• Remote Procedure Call for communication among
processors
– DCOM, CORBA, Java RMI, etc.
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
26
Shared-Memory Multiprocessor
• All processors can access all memory
• Some memory may be faster than other memory
• Any process can run on any processor
• Multiple processors executing in same address space
concurrently
• Multiple paths to disks (as with cluster)
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
27
Shared Memory Multiprocessors
Bus-based
•Bus contention limits
the number of CPUs
CS-502 Fall 2006
•Lower bus contention
•Caches need to be
synced (big deal)
Multiprocessing and
Distributed Systems
•Compiler places data
and text in private or
shared memory
28
Multiprocessors (2) - Crossbar
•
•
•
Can support a large number of CPUs Non-blocking network
Cost/performance effective up to about 100 CPU – growing as n2
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
29
Multiprocessors(3) – Multistage Switching Networks
• Omega Network – blocking
– Lower cost, longer latency
– For N CPUs and N memories – log2 n stages of n/2 switches
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
30
Type of Multiprocessors – UMA vs. NUMA
• UMA (Uniform
Memory Access)
• NUMA (Non-Uniform
Memory Access)
– All memory is equal
– Familiar programming
model
– Number of processors
is limited
• 8-32
– Symmetrical
– Single address space
visible to all CPUs
– Access to remote
memory via commands
• LOAD & STORE
• remote memory access
slower than to local
• No processor is
different from others
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
31
Common SMP Implementations
• Multiple processors on same board or bus
• Dual core processors
• I.e., two processors on same chip
• Share same L2 cache
• Hyperthreading
• One processor shares its circuitry among two
threads or processes
• Separate registers, PSW, etc.
• Combinations of above
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
32
Hardware issue – Cache Synchronization
• Each processor cache monitors memory
accesses on bus (i.e., “snoops”)
– Memory updates are propagated to other caches
– Complex cache management hardware
– Order of operations may be ambiguous
• Some data must be marked as not cacheable
– Visible to programming model
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
33
Operating System Considerations
• Master-slave vs. full symmetry
• Explicit cache flushing
• When page is replaced or invalidated
• Processor affinity of processes
• Context switches across processors can be
expensive
• Synchronization across processors
• Interrupt handling
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
34
Multiprocessor OS – Master-Slave
• One CPU (master) runs the OS and applies
most policies
• Other CPUs
– run applications
– Minimal OS to acquire and terminate processes
• Relatively simple OS
• Master processor can become a bottleneck
for a large number of slave processors
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
35
Multiprocessor OS –
Symmetric Multi-Processor (SMP)
• Any processor can execute the OS and any
applications
• Synchronization within the OS is the issue
1. Lock the whole OS – poor utilization – long queues
waiting to use OS
2. OS critical regions – much preferred
– Identify independent OS critical regions that be executed
independently – protect with mutex
– Identify independent critical OS tables – protect access with
MUTEX
– Design OS code to avoid deadlocks
–
–
CS-502 Fall 2006
The art of the OS designer
Maintenance requires great care
Multiprocessing and
Distributed Systems
36
Multiprocessor OS – SMP (continued)
• Multiprocessor Synchronization
– Need special instructions – test-and-set
– Spinlocks are common
• Can context switch if time in critical region is
greater than context switch time
• OS designer must understand the performance of OS
critical regions
• Context switch time could be onerous
– Data cached on one processor needs to be recached on another
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
37
Multiprocessor Scheduling
• When processes are independent (e.g., timesharing)
– Allocate CPU to highest priority process
– Tweaks
• For a process with a spinlock, let it run until it releases the lock
• To reduce TLB and memory cache flushes, try to run a process
on the same CPU each time it runs
• For groups of related processes
– Attempt to simultaneously allocate CPUs to all related
processes (space sharing)
– Run all threads to termination or block
– Gang schedule – apply a scheduling policy to related
processes together
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
38
Linux SMP Support
• Multi-threaded kernel
• Any processor can handle any interrupt
• Interrupts rarely disabled, and only for shortest time possible
• Multiple processors can be active in system calls concurrently
• Process vs. Interrupt context
• Process context:– when processor is acting on behalf of a
specific process
• Interrupt context:– when processor is acting on behalf of no
particular process
• Extensive use of spinlocks
• Processor affinity properties in task_struct
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
39
WPI Computing & Communications Center
•
•
•
•
•
•
CCC cluster – 11 dual 2.4 GHz Xeon processors
toth – 16-node 1.5 GHz Itanium2
big16 – 8 dual-core 2.4 GHz Opterons
toaster – NetApp 5.5 TByte Fiber Channel RAID NFS
breadbox – NetApp 12 TByte RAID NFS
Many others
• All connected together by very fast networks
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
40
Questions?
Next Topic
CS-502 Fall 2006
Multiprocessing and
Distributed Systems
41