편집토론문서 역사

Latency 101: 두 판 사이의 차이

 
(같은 사용자의 중간 판 하나는 보이지 않습니다)
7번째 줄: 7번째 줄:
L2 cache reference                          7  ns                      14x L1 cache
L2 cache reference                          7  ns                      14x L1 cache
Mutex lock/unlock                          25  ns         
Mutex lock/unlock                          25  ns         
VMFUNC with eptp switching                  54  ns
Main memory reference                      100  ns                      20x L2 cache, 200x L1 cache
Main memory reference                      100  ns                      20x L2 cache, 200x L1 cache
getpid                                    120  ns
getpid                                    120  ns
CR3 Update without TLB flush              120  ns
CR3 Update without TLB flush              120  ns
VMFUNC with eptp switching                120  ns
Compress 1K bytes with Zippy            3,000  ns        3 us
Compress 1K bytes with Zippy            3,000  ns        3 us
Send 1K bytes over 1 Gbps network      10,000  ns      10 us
Send 1K bytes over 1 Gbps network      10,000  ns      10 us
35번째 줄: 35번째 줄:
WRPKRU                                      22  cycles
WRPKRU                                      22  cycles
RDPKRU                                        7  cycles
RDPKRU                                        7  cycles
WRMSR (0x6E1 PKS)                          320  cycles
RDMSR (0x6E1 PKS)                          120  cycles
</syntaxhighlight>
</syntaxhighlight>


2023년 6월 2일 (금) 07:42 기준 최신판


Stroage Latency Comparison

L1 cache reference                           0.5 ns
Branch mispredict                            5   ns
L2 cache reference                           7   ns                      14x L1 cache
Mutex lock/unlock                           25   ns        
VMFUNC with eptp switching                  54   ns
Main memory reference                      100   ns                      20x L2 cache, 200x L1 cache
getpid                                     120   ns
CR3 Update without TLB flush               120   ns
Compress 1K bytes with Zippy             3,000   ns        3 us
Send 1K bytes over 1 Gbps network       10,000   ns       10 us
Read 4K randomly from SSD*             150,000   ns      150 us          ~1GB/sec SSD
Read 1 MB sequentially from memory     250,000   ns      250 us
Round trip within same datacenter      500,000   ns      500 us
Read 1 MB sequentially from SSD*     1,000,000   ns    1,000 us    1 ms  ~1GB/sec SSD, 4X memory
Disk seek                           10,000,000   ns   10,000 us   10 ms  20x datacenter roundtrip
Read 1 MB sequentially from disk    20,000,000   ns   20,000 us   20 ms  80x memory, 20X SSD
Send packet CA->Netherlands->CA    150,000,000   ns  150,000 us  150 ms

Instructions

MOV                                           1   cycles
JMP (Short)                                   1   cycles
JMP (Far)                                 16-20   cycles
CALL (Far)                                16-22   cycles
CALL (Near)                                   3   cycles
INT                                          33   cycles
IRET                                         32   cycles
CLI                                         8-9   cycles
STI                                       16-17   cycles
RDTSC                                         5   cycles
WRPKRU                                       22   cycles
RDPKRU                                        7   cycles
WRMSR (0x6E1 PKS)                           320   cycles
RDMSR (0x6E1 PKS)                           120   cycles

참고

  1. https://gist.github.com/jboner/2841832/forks
  2. https://www.agner.org/optimize/instruction_tables.pdf