Term
|
Definition
Main Memory
ALU
Program Control
I/O Equipment |
|
|
Term
|
Definition
Transistors doubling on chip yearly
Later revised to every 18 months |
|
|
Term
|
Definition
| Before executing a program, all instructions are changed to lower level language and then executed |
|
|
Term
|
Definition
| A program accepts L1 instructions and executes them as L0 when it receives the. It doesn't require first generating a program in L0. |
|
|
Term
|
Definition
| A CPU component which directs operation of the processor. |
|
|
Term
|
Definition
| A small amount of storage on a processor. Faster than RAM. |
|
|
Term
|
Definition
| Start, Fetch, Execute, Loop, Stop |
|
|
Term
| Benefits of Computers with Interpreted Instructions: |
|
Definition
fix and add instructions
efficient testing of complex instructions |
|
|
Term
| arithmetic and logic unit (ALU) |
|
Definition
| A digital circuit that performs integer and logical operations |
|
|
Term
|
Definition
Control Unit
ALU – Arithmetic Logic Unit
Registers |
|
|
Term
| Design Principles for Modern Computers |
|
Definition
All instructions are directly executed by hardware
Maximize the rate at which instructions are issued
Instructions should be easy to decode
Only loads and stores should reference memory
Provide plenty of registers |
|
|
Term
|
Definition
Processor – registers
Internal – main memory or RAM
External – tape, secondary storage |
|
|
Term
|
Definition
| A CPU cache is a cache used by the central processing unit of a computer to reduce the average time to access memory. The cache is a smaller, faster memory which stores copies of the data from frequently used main memory locations. As long as most memory accesses are cached memory locations, the average latency of memory accesses will be closer to the cache latency than to the latency of main memory. |
|
|
Term
|
Definition
| A track is a series of sectors, on the same physical platter of a disk, all of which can be accessed without moving the disk's read/write heads. |
|
|
Term
|
Definition
| A sector or group of sectors |
|
|
Term
|
Definition
This is the smallest unit of storage on the disk which the system can access, and is typically 512 bytes.
Area between two radius and the arc (outside of disk) |
|
|
Term
|
Definition
| A cylinder is the set of tracks, each of which is on a different platter and all of which can be accessed without moving the disks's read/write heads. When a disk is divided into different logical partitions (or slices), these are specified in terms of the start cylinder of each partition plus the number of cylinders it contains. Typically some small number of cylinders is reserved i.e. not available to the system for use within disk partitions. |
|
|
Term
|
Definition
| Disk-drive manufacturers also offered fixed-head drives. In these, the heads did not move back and forth. Instead of mechanically scanning the disk, the drive electronically selected different heads, speeding up data access significantly. This design required several heads to cover as much of the disk surface as possible. |
|
|
Term
| Magnetic Disk moveable head |
|
Definition
| Head moves toward the center (forward)and outside (backward) of the disk while reading information |
|
|
Term
| Magnetic Disk removable/ non-removable |
|
Definition
removable you can take it out of the computer.
non removable you can't easily take it out of the computer. |
|
|
Term
| Magnetic Disk single/double-sided |
|
Definition
| data is stored on only one side versus two sides |
|
|
Term
| Magnetic Disk Single-platter/multiple-platter |
|
Definition
| Disks are made up of a number of circular platters, each of which is has a magnetic coating. The platters are stacked one on top of another and rotate together round a central spindle. Data is read from and written to these using a number of read/write heads each of which is attached to an arm which can be moved in order to access different parts of the disk. |
|
|
Term
|
Definition
a.set of physical drives system sees as 1 drive
b. data distributed across physical drives in an array
c. parity information – data recovery (not RAID 0) |
|
|
Term
|
Definition
Striping: non-redundant, high performance read/write
a1 a2
a3 a4
a5 a6
a7 a8 |
|
|
Term
|
Definition
Mirroring: system critical files/drives
a1 a1
a2 a2
a3 a3
a4 a4 |
|
|
Term
|
Definition
parity across all disk, high read speed, slightly slower write speed, minimum of 3
a1 a2 a3 ap
b1 b2 bp b3
c1 cp c2 c3
dp d1 d2 d3 |
|
|
Term
|
Definition
| constant angular velocity - hard disks - disk spins at a set rate of rpms at all times |
|
|
Term
|
Definition
| constant linear velocity - CD ROMS, BLURAY, DVD - speed increases as you reader farther from center |
|
|
Term
|
Definition
| A communication system that transfers data between components inside a computer, or between computers. |
|
|
Term
|
Definition
| monitors, modems, keyboards |
|
|
Term
|
Definition
| The use of geometrical primitives such as points, lines, curves, and shapes or polygon(s), which are all based on mathematical expressions, to represent images in computer graphics. |
|
|
Term
| Raster graphics image/bitmap |
|
Definition
| A dot matrix data structure representing a generally rectangular grid of pixels, or points of color, view-able via a monitor, paper, or other display medium. |
|
|
Term
|
Definition
Static RAM does not have to be periodically refreshed because it uses switches rather than charges.
SRAM used for a computer's cache memory |
|
|
Term
|
Definition
Dynamic - storage cells refreshed or given a new electronic charge every few milliseconds.
Used for main memory
One transistor and a capacitor are required per bit, compared to four or six transistors in SRAM. This allows DRAM to reach very high densities. |
|
|
Term
|
Definition
| Synchronized with the system bus - allowing greater efficiency in storing and retrieving data. Can send more at once. |
|
|
Term
|
Definition
| A programmable read-only memory is a form of digital memory where the setting of each bit is locked by a fuse or antifuse. Such PROMs are used to store programs permanently. |
|
|
Term
|
Definition
| EPROM (erasable and programmable ROM)-can be programmed and then erased (typically erased with a UV [ultra-violet light] eraser). These chips are easy to spot because they have a "window" so they can be erased. |
|
|
Term
|
Definition
| Same as EPROM but is erased electrically. These don't have a "window". |
|
|
Term
|
Definition
| Sequencing unrelated activities such that they use different components at the same time. By keeping as much of the different components active at once you maximize efficiency and speed up execution time. |
|
|
Term
| Parallelism - Superscalar |
|
Definition
| Multiple sub-components of the CPU capable of doing the same task simultaneously, but with the processor deciding how to do it. |
|
|
Term
|
Definition
| one of the memory-management schemes by which a computer can store and retrieve data from secondary storage for use in main memory - such as using space on a hard drive for RAM. |
|
|
Term
|
Definition
| There isn't enough free RAM so the computer has to constantly exchange RAM info for page file info. This greatly slows down the computer's speed. |
|
|
Term
|
Definition
| Have the chip do more things at the same time to increase throughput. |
|
|
Term
|
Definition
| Allows the CPU to manage multiple threads of control at the same time in an attempt to be more efficient. If thread is waiting on next word to make it to cache then the cpu will start working on another thread until it's available. |
|
|
Term
| Single Chip Multi-processors |
|
Definition
| two or more cpus on the same chip - multicore |
|
|
Term
|
Definition
| A second, specialized processor - for example floating point or graphics that the main cpu can hand of instructions to. |
|
|
Term
| Shared Memory Multi Processors |
|
Definition
| A parallel computer in which all cpus share a common memory. They can communicate by having one of them write data to memory and the other one read that data |
|
|
Term
| Symetric MultiProcessor Systems |
|
Definition
| multiprocessor where every processor has access to all i/o devices |
|
|
Term
|
Definition
| how links and switches are arranged in a network - star, tree, 4d hypercube, grid, ring, double torus, cube |
|
|
Term
| Massively Parallel Processor |
|
Definition
| Supercomputer made of many cpus coupled on network. |
|
|
Term
|
Definition
| A computer cluster consists of a set of loosely connected or tightly connected computers that work together so that in many respects they can be viewed as a single system. |
|
|
Term
|
Definition
| Grid computing is the collection of computer resources from multiple locations to reach a common goal. |
|
|
Term
| Evolution of I/O (Handout) |
|
Definition
1. CPU directly controls peripheral
2. Controller or I/O is added; CPU uses programmed I/O without interrupts
3. Same as 2 but now interrupts are used freeing the CPU
4. DMA – I/O has direct access to memory
5. I/O module is now a processor in its own right and has its own instruction set
6. Same as 5 but now I/O has its own local memory – (I/O as a computer) |
|
|
Term
|
Definition
1. Reduce the number of clock cycles needed to execute an instruction
– Reduce the Path Length: the number of clock cycles needed to execute a set of operations is known as the path length.
– Merge the interpreter loop into the end of each microcode sequence
2. Simplify the organization so that the clock cycle can be shorter
– Go from a 2 bus design to a 3 bus design to allow register to register addition in one cycle.
– have instructions fetched from memory by a specialized functional unit (instruction pre-fetch unit)
3. Overlap the execution of instructions
– Pipelining -Insert latches on each bus so that it: a. speeds up the clock cycle b. Uses all parts of the data path during every cycle. |
|
|
Term
| Improving Performance 4.5 |
|
Definition
1. Cache Memory
2. Branch Prediction
3. Out-of-Order Execution and Register Re-naming
4. Speculative Execution |
|
|
Term
|
Definition
| Branch predictor is a digital circuit that tries to guess which way a branch (e.g. an if-then-else structure) will go before this is known for sure. The purpose of the branch predictor is to improve the flow in the instruction pipeline. |
|
|
Term
|
Definition
| Random access means that the PC processor can access any part of the memory directly rather than having to proceed sequentially from some starting place. |
|
|
Term
|
Definition
| The more address lines a bus has the more memory the CPU can address directly. |
|
|
Term
|
Definition
| Processors, Co-processors and I/O Controller chips negotiate communications between each other based on assigned priorities |
|
|
Term
|
Definition
| Speculative execution is an optimization technique where a computer system performs some task that may not be actually needed. The main idea is to do work before it is known whether that work will be needed at all, so as to prevent a delay that would have to be incurred by doing the work after it is known whether it is needed. If it turns out the work was not needed after all, any changes made by the work are reverted and the results are ignored. |
|
|
Term
|
Definition
A word is basically a fixed-sized group of digits (binary or decimal) that are handled as a unit by the instruction set and/or hardware of the processor.
d-word 32 bits q-word 64 bits |
|
|
Term
|
Definition
| In systems programming, an interrupt is a signal to the processor emitted by hardware or software indicating an event that needs immediate attention. |
|
|
