H13-211 HCIA-Intelligent Computing V1.0 Questions and Answers

GPU has fewer execution units and can only process instructions in a single thread

The advantage of GPU is that if the computer needs to be changed, it can be reassembled

FPGA can be based on specific application cloud programming hardware, flexibility is much better than GPU

For multi-peak performance, GPU is lower than PPGA

TRUE

FALS

Applicable graphics and video processing

General processor, the shortest development cycle

Suitable for irregular parallel computing, good real-time performance, high energy efficiency ratio

Suitable for in-depth customization and can be launched quickly

The middleware provides the connection software between the system software and the application software

The system software is responsible for managing various independent hardware in the computer system, providing a platform for the running application software, and adapting to different application software

Application software is software developed for a specific purpose

Computer software divides system software, application software and middleware between them

Process

Register

Control unit

Logic operation unit

Application system

Business application layer system

Data logic system

Number estimation

Processor optimization package/physical IP package authorization

Processor authorization

Architecture/instruction set authorization

CPU

GPU

FPGA

ASIC

Manageability

reliability

Usability

Protectability

TRUE

FALSE

Applications with frequent random reads and writes, such as OLTP services. IOPS is a key measurement indicator. Another important indicator is data throughput, which refers to the amount of data that can be successfully transferred per unit time

The hard disk data transfer rate refers to the speed at which the hard disk reads and writes data, and the unit is knots per second (MB/S)

The factors affecting the capacity of the hard disk are the capacity of a single disc and the number of discs

The average access time of the mechanical hard disk refers to the average addressing time of the disk

Throughput and error rate

Throughput and latency

Transmission rate, error rate

Cluster server added to the cluster system

Increase the memory capacity of a single server in the cluster

Increase the networking bandwidth in the cluster

Replace the hard disk of the single server in the cluster with a solid-state hard disk with a height of 10

Under the same external frequency, the higher the frequency multiplier is, the higher the thing is, without restrictions

The speed of data transmission between the CPU and the system is limited, and the CPU that blindly pursues high principle and obtains high frequency multiplication will have a bottleneck effect of the net legs.

The frequency doubling system refers to the relative proportion between the main CPU item and Waiton

R Guo Zuo is right to adjust the signal frequency of the CFU without indignation. The super state that adjusts the frequency multiplier is quite stable.

TRUE

FALSE

Increase the networking bandwidth in the cluster

Increase the memory capacity of a single server in the cluster

Add a new cluster server to the cluster system

Replace the hard disk of a single server in the cluster with a high-IO solid-state disk

Memory is relative to the outside world, its main feature is fast access rate

Memory is also called internal storage, its role is to backup and copy and permanently store the calculation data in the CPU to improve the reliability of the data

Memory, also known as main memory, is a storage space that the CPU can directly address and is made of semiconductor devices

When the computer is running, the CPU will transfer the data that needs to be calculated to the memory for operation. When the operation is completed, the CPU will transfer the results out. The operation of the memory also determines the stable operation of the computer.

16X

1X

8X

4X

LVS has three working modes LVS-NAT (address translation), LVS-DR (direct routing), LVS-TUN (tunnel)

LVS commonly used scheduling strategies are, polling scheduling, weighted polling scheduling, minimum number of connections, fast response priority, etc.

LVS is a virtual server cluster, working at the seventh layer of the network system, using IP load balancing technology and the fourth layer based on content request distribution technology

LVS cluster is mainly composed of three parts: load scheduler and server pool (server pool/Reslserver) shared storage (shared storage)

TRUE

FALSE

The computing node only provides computing resources such as CPU and memory. The hardware configuration information is stored in the configuration file, and the data is stored centrally on the disk array, which realizes the independence of the network, storage and computing node.

The computing node configuration is saved in the chassis management module, which can be imported and exported. The chassis management module's master and backup are automatically synchronized, and the master and backup are mutually backup.

Stateless computing technology eliminates the dependence on the hypervisor in the virtualization layer, so the database can use the hardware-based stateless computing technology to achieve a low-cost high availability solution