Compaq 108164-003 - ProLiant - 800 Integration Notes page 11

ECG045/1097
I N
NTEGRATION OTE
(cont.)
. .
. .
. .
G M
. . L O S S A R Y O F
. .
. .
. . Bandwidth
. .
Increasing the network operating system server capacity by reducing CPU utilization and
. .
allowing more services and applications to run.
. .
. .
. .
Cache Coherency
. .
. .
The monitoring of the state of data stored in multiple caches to ensure that no processor acts on
. .
stale data. Locks: Controls within multithreaded software that prevent more than one thread from
. .
. .
accessing the same area of memory simultaneously.
. .
. .
. .
Multitasking
. .
The execution of multiple applications or threads to provide simultaneous execution.
. .
. .
. .
Multithreading
. .
. .
The multitasking of separate threads within a process or application.
. .
. .
. .
Mutex
. .
Mutual Exclusion refers to atomic lock variables which are used to prevent multiple processors
. .
from accessing shared data structures at the same instant.
. .
. .
. .
Platform Specific Module (PSM)
. .
. .
A driver or layer that interfaces between the operating system and the hardware, freeing the OS
. .
from being dependent upon particular hardware components.
. .
. .
. .
Process
. .
. .
One or more threads and the resources allocated to them, including a virtual address space. Each
. .
instance of any application runs within its own process.
. .
. .
. .
Processor Affinity
. .
. .
The assignment of threads to a given CPU or subset of CPUs to capitalize on processor caches. A
. .
high percentage of cache hits results in improved performance.
. .
. .
. .
Scaling
. .
. . The percentage of increase in throughput, or work accomplished--relative to processor 0--when
. .
additional processors are added. Scalability is the incremental increased related to the addition of
. .
. . each processor.
. .
. .
Semaphore
. .
. .
Restricts access to global variables or system-level resources. For example, it blocks additional
. .
threads from writing to a serial port while another thread is using it. When a thread obtains a
. .
. .
semaphore, it changes the resource's state to owned, then when completed, changes the
. .
state back to unowned and releases the semaphore.
. .
. .
. .
Symmetric Multiprocessing (SMP)
. .
. .
The ability of an operating system to distribute tasks to all available CPUs in a hardware platform
. .
and to execute these tasks in parallel.
. .
. .
. .
Thread
. .
. .
A single path or entity of execution within one process.
. .
. .
. .
. .
11
U L T I P R O C E S S I N G
T
E R M I N O L O G Y