C語言字串壓縮之ZSTD演演算法詳解

2022-08-10 22:00:34

前言

最近專案上有大量的字串資料需要儲存到記憶體，並且需要儲存至一定時間，於是自然而然的想到了使用字串壓縮演演算法對“源串”進行壓縮儲存。由此觸發了對一些優秀壓縮演演算法的調研。

字串壓縮，我們通常的需求有幾個，一是高壓縮率，二是壓縮速率高，三是解壓速率高。不過高壓縮率與高壓縮速率是魚和熊掌的關係，不可皆得，優秀的演演算法一般也是採用壓縮率與效能折中的方案。從壓縮率、壓縮速率、解壓速率考慮，zstd與lz4有較好的壓縮與解壓效能，最終選取zstd與lz4進行調研。

zstd是facebook開源的提供高壓縮比的快速壓縮演演算法（參考https://github.com/facebook/zstd），很想了解一下它在壓縮與解壓方面的實際表現。

一、zstd壓縮與解壓

ZSTD_compress屬於ZSTD的Simple API範疇，只有壓縮級別可以設定。

ZSTD_compress函數原型如下：

size_t ZSTD_compress(void* dst, size_t dstCapacity, const void* src, size_t srcSize, int compressionLevel)

ZSTD_decompress函數原型如下：

size_t ZSTD_decompress( void* dst, size_t dstCapacity, const void* src, size_t compressedSize);　　我們先來看看zstd的壓縮與解壓縮範例。

#include <stdio.h>
#include <string.h>
#include <sys/time.h>
#include <malloc.h>
#include <zstd.h>
#include <iostream>

using namespace std;

int main()
{
    // compress
    size_t com_space_size;
    size_t peppa_pig_text_size;

    char *com_ptr = NULL;
    char peppa_pig_buf[2048] = "Narrator: It is raining today. So, Peppa and George cannot play outside.Peppa: Daddy, it's stopped raining. Can we go out to play?Daddy: Alright, run along you two.Narrator: Peppa loves jumping in muddy puddles.Peppa: I love muddy puddles.Mummy: Peppa. If you jumping in muddy puddles, you must wear your boots.Peppa: Sorry, Mummy.Narrator: George likes to jump in muddy puddles, too.Peppa: George. If you jump in muddy puddles, you must wear your boots.Narrator: Peppa likes to look after her little brother, George.Peppa: George, let's find some more pud dles.Narrator: Peppa and George are having a lot of fun. Peppa has found a lttle puddle. George hasfound a big puddle.Peppa: Look, George. There's a really big puddle.Narrator: George wants to jump into the big puddle first.Peppa: Stop, George. | must check if it's safe for you. Good. It is safe for you. Sorry, George. It'sonly mud.Narrator: Peppa and George love jumping in muddy puddles.Peppa: Come on, George. Let's go and show Daddy.Daddy: Goodness me.Peppa: Daddy. Daddy. Guess what we' ve been doing.Daddy: Let me think... Have you been wa tching television?Peppa: No. No. Daddy.Daddy: Have you just had a bath?Peppa: No. No.Daddy: | know. You've been jumping in muddy puddles.Peppa: Yes. Yes. Daddy. We've been jumping in muddy puddles.Daddy: Ho. Ho. And look at the mess you're in.Peppa: Oooh....Daddy: Oh, well, it's only mud. Let's clean up quickly before Mummy sees the mess.Peppa: Daddy, when we've cleaned up, will you and Mummy Come and play, too?Daddy: Yes, we can all play in the garden.Narrator: Peppa and George are wearing their boots. Mummy and Daddy are wearing their boots.Peppa loves jumping up and down in muddy puddles. Everyone loves jumping up and down inmuddy puddles.Mummy: Oh, Daddy pig, look at the mess you're in. .Peppa: It's only mud.";

    peppa_pig_text_size = strlen(peppa_pig_buf);
    com_space_size= ZSTD_compressBound(peppa_pig_text_size);
    com_ptr = (char *)malloc(com_space_size);
    if(NULL == com_ptr) {
        cout << "compress malloc failed" << endl;
        return -1;
    }

    size_t com_size;
    com_size = ZSTD_compress(com_ptr, com_space_size, peppa_pig_buf, peppa_pig_text_size, ZSTD_fast);
    cout << "peppa pig text size:" << peppa_pig_text_size << endl;
    cout << "compress text size:" << com_size << endl;
    cout << "compress ratio:" << (float)peppa_pig_text_size / (float)com_size << endl << endl;


    // decompress
    char* decom_ptr = NULL;
    unsigned long long decom_buf_size;
    decom_buf_size = ZSTD_getFrameContentSize(com_ptr, com_size);

    decom_ptr = (char *)malloc((size_t)decom_buf_size);
    if(NULL == decom_ptr) {
        cout << "decompress malloc failed" << endl;
        return -1;
    }

    size_t decom_size;
    decom_size = ZSTD_decompress(decom_ptr, decom_buf_size, com_ptr, com_size);
    cout << "decompress text size:" << decom_size << endl;

    if(strncmp(peppa_pig_buf, decom_ptr, peppa_pig_text_size)) {
        cout << "decompress text is not equal peppa pig text" << endl;
    }

    free(com_ptr);
    free(decom_ptr);
    return 0;
}

執行結果：

從結果可以發現，壓縮之前的peppa pig文字長度為1827，壓縮後的文字長度為759，壓縮率為2.4，解壓後的長度與壓縮前相等。

另外，上文提到可以調整ZSTD_compress函數的壓縮級別，zstd的預設級別為ZSTD_CLEVEL_DEFAULT = 3，最小值為0，最大值為ZSTD_MAX_CLEVEL = 22。另外也提供一些策略設定，例如 ZSTD_fast, ZSTD_greedy, ZSTD_lazy, ZSTD_lazy2, ZSTD_btlazy2。壓縮級別越高，壓縮率越高，但是壓縮速率越低。

二、ZSTD壓縮與解壓效能探索

上面探索了zstd的基礎壓縮與解壓方法，接下來再摸索一下zstd的壓縮與解壓縮效能。

測試方法是，使用ZSTD_compress連續壓縮同一段文字並持續10秒，最後得到每一秒的平均壓縮速率。測試壓縮效能的程式碼範例如下：

#include <stdio.h>
#include <string.h>
#include <sys/time.h>
#include <malloc.h>
#include <zstd.h>
#include <iostream>

using namespace std;

int main()
{
    int cnt = 0;

    size_t com_size;
    size_t com_space_size;
    size_t peppa_pig_text_size;

    char *com_ptr = NULL;
    char peppa_pig_buf[2048] = "Narrator: It is raining today. So, Peppa and George cannot play outside.Peppa: Daddy, it's stopped raining. Can we go out to play?Daddy: Alright, run along you two.Narrator: Peppa loves jumping in muddy puddles.Peppa: I love muddy puddles.Mummy: Peppa. If you jumping in muddy puddles, you must wear your boots.Peppa: Sorry, Mummy.Narrator: George likes to jump in muddy puddles, too.Peppa: George. If you jump in muddy puddles, you must wear your boots.Narrator: Peppa likes to look after her little brother, George.Peppa: George, let's find some more pud dles.Narrator: Peppa and George are having a lot of fun. Peppa has found a lttle puddle. George hasfound a big puddle.Peppa: Look, George. There's a really big puddle.Narrator: George wants to jump into the big puddle first.Peppa: Stop, George. | must check if it's safe for you. Good. It is safe for you. Sorry, George. It'sonly mud.Narrator: Peppa and George love jumping in muddy puddles.Peppa: Come on, George. Let's go and show Daddy.Daddy: Goodness me.Peppa: Daddy. Daddy. Guess what we' ve been doing.Daddy: Let me think... Have you been wa tching television?Peppa: No. No. Daddy.Daddy: Have you just had a bath?Peppa: No. No.Daddy: | know. You've been jumping in muddy puddles.Peppa: Yes. Yes. Daddy. We've been jumping in muddy puddles.Daddy: Ho. Ho. And look at the mess you're in.Peppa: Oooh....Daddy: Oh, well, it's only mud. Let's clean up quickly before Mummy sees the mess.Peppa: Daddy, when we've cleaned up, will you and Mummy Come and play, too?Daddy: Yes, we can all play in the garden.Narrator: Peppa and George are wearing their boots. Mummy and Daddy are wearing their boots.Peppa loves jumping up and down in muddy puddles. Everyone loves jumping up and down inmuddy puddles.Mummy: Oh, Daddy pig, look at the mess you're in. .Peppa: It's only mud.";

    timeval st, et;

    peppa_pig_text_size = strlen(peppa_pig_buf);
    com_space_size= ZSTD_compressBound(peppa_pig_text_size);

    gettimeofday(&st, NULL);
    while(1) {

        com_ptr = (char *)malloc(com_space_size);
        com_size = ZSTD_compress(com_ptr, com_space_size, peppa_pig_buf, peppa_pig_text_size, ZSTD_fast);

        free(com_ptr);
        cnt++;

        gettimeofday(&et, NULL);
        if(et.tv_sec - st.tv_sec >= 10) {
            break;
        }
    }

    cout << "compress per second:" << cnt/10 << " times" << endl;
    return 0;
}

執行結果：

結果顯示ZSTD的壓縮效能大概在每秒6-7萬次左右，這個結果其實並不是太理想。需要說明的是壓縮效能與待壓縮文字的長度、字元內容也是有關係的。

我們再來探索一下ZSTD的解壓縮效能。與上面的測試方法類似，先對本文進行壓縮，然後連續解壓同一段被壓縮過的資料並持續10秒，最後得到每一秒的平均解壓速率。測試解壓效能的程式碼範例如下：

#include <stdio.h>
#include <string.h>
#include <sys/time.h>
#include <malloc.h>
#include <zstd.h>
#include <iostream>

using namespace std;

int main()
{
    int cnt = 0;

    size_t com_size;
    size_t com_space_size;
    size_t peppa_pig_text_size;

    timeval st, et;

    char *com_ptr = NULL;
    char peppa_pig_buf[2048] = "Narrator: It is raining today. So, Peppa and George cannot play outside.Peppa: Daddy, it's stopped raining. Can we go out to play?Daddy: Alright, run along you two.Narrator: Peppa loves jumping in muddy puddles.Peppa: I love muddy puddles.Mummy: Peppa. If you jumping in muddy puddles, you must wear your boots.Peppa: Sorry, Mummy.Narrator: George likes to jump in muddy puddles, too.Peppa: George. If you jump in muddy puddles, you must wear your boots.Narrator: Peppa likes to look after her little brother, George.Peppa: George, let's find some more pud dles.Narrator: Peppa and George are having a lot of fun. Peppa has found a lttle puddle. George hasfound a big puddle.Peppa: Look, George. There's a really big puddle.Narrator: George wants to jump into the big puddle first.Peppa: Stop, George. | must check if it's safe for you. Good. It is safe for you. Sorry, George. It'sonly mud.Narrator: Peppa and George love jumping in muddy puddles.Peppa: Come on, George. Let's go and show Daddy.Daddy: Goodness me.Peppa: Daddy. Daddy. Guess what we' ve been doing.Daddy: Let me think... Have you been wa tching television?Peppa: No. No. Daddy.Daddy: Have you just had a bath?Peppa: No. No.Daddy: | know. You've been jumping in muddy puddles.Peppa: Yes. Yes. Daddy. We've been jumping in muddy puddles.Daddy: Ho. Ho. And look at the mess you're in.Peppa: Oooh....Daddy: Oh, well, it's only mud. Let's clean up quickly before Mummy sees the mess.Peppa: Daddy, when we've cleaned up, will you and Mummy Come and play, too?Daddy: Yes, we can all play in the garden.Narrator: Peppa and George are wearing their boots. Mummy and Daddy are wearing their boots.Peppa loves jumping up and down in muddy puddles. Everyone loves jumping up and down inmuddy puddles.Mummy: Oh, Daddy pig, look at the mess you're in. .Peppa: It's only mud.";

    size_t decom_size;
    char* decom_ptr = NULL;
    unsigned long long decom_buf_size;

    peppa_pig_text_size = strlen(peppa_pig_buf);
    com_space_size= ZSTD_compressBound(peppa_pig_text_size);
    com_ptr = (char *)malloc(com_space_size);

    com_size = ZSTD_compress(com_ptr, com_space_size, peppa_pig_buf, peppa_pig_text_size, 1);

    gettimeofday(&st, NULL);
    decom_buf_size = ZSTD_getFrameContentSize(com_ptr, com_size);

    while(1) {

        decom_ptr = (char *)malloc((size_t)decom_buf_size);

        decom_size = ZSTD_decompress(decom_ptr, decom_buf_size, com_ptr, com_size);
        if(decom_size != peppa_pig_text_size) {

            cout << "decompress error" << endl;
            break;
        }

        free(decom_ptr);

        cnt++;
        gettimeofday(&et, NULL);
        if(et.tv_sec - st.tv_sec >= 10) {
            break;
        }
    }

    cout << "decompress per second:" << cnt/10 << " times" << endl;

    free(com_ptr);
    return 0;
}

執行結果：

結果顯示ZSTD的解壓縮效能大概在每秒12萬次左右，解壓效能比壓縮效能高。

三、zstd的高階用法

zstd提供了一個名為PZSTD的壓縮和解壓工具。PZSTD（parallel zstd），並行壓縮的zstd，是一個使用多執行緒對待壓縮文字進行切片分段，且進行並行壓縮的命令列工具。

其實高版本（v1.4.0及以上）的zstd也提供了指定多執行緒對文字進行並行壓縮的相關API介面，也就是本小節要介紹的zstd高階API用法。下面我們再來探索一下zstd的多執行緒壓縮使用方法。

多執行緒並行壓縮的兩個關鍵API，一個是引數設定API，另一個是壓縮API。

引數設定API的原型是：

size_t ZSTD_CCtx_setParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int value)

壓縮API的原型是：

size_t ZSTD_compress2(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize)

下面給出zstd並行壓縮的範例demo，通過ZSTD_CCtx_setParameter設定執行緒數為3，即指定宏ZSTD_c_nbWorkers為3，通過ZSTD_compress2壓縮相關文字。另外，為了展示zstd確實使用了多執行緒，需要先讀取一個非常大的檔案，作為zstd的壓縮文字源，儘量使zstd執行較長時間。

#include <stdio.h>
#include <string.h>
#include <sys/time.h>
#include <malloc.h>
#include <zstd.h>
#include <iostream>

using namespace std;

int main()
{
    size_t com_size;
    size_t com_space_size;

    FILE *fp = NULL;
    unsigned int file_len;

    char *com_ptr = NULL;
    char *file_text_ptr = NULL;

    fp = fopen("xxxxxx", "r");
    if(NULL == fp){
         cout << "file open failed" << endl;
         return -1;
    }

    fseek(fp, 0, SEEK_END);
    file_len = ftell(fp);
    fseek(fp, 0, SEEK_SET);
    cout << "file length:" << file_len << endl;

    // malloc space for file content
    file_text_ptr = (char *)malloc(file_len);
    if(NULL == file_text_ptr) {
        cout << "malloc failed" << endl;
        return -1;
    }

    // malloc space for compress space
    com_space_size = ZSTD_compressBound(file_len);
    com_ptr = (char *)malloc(com_space_size);
    if(NULL == com_ptr) {
        cout << "malloc failed" << endl;
        return -1;
    }

    // read text from source file
    fread(file_text_ptr, 1, file_len, fp);
    fclose(fp);

    ZSTD_CCtx* cctx;
    cctx = ZSTD_createCCtx();

    // set multi-thread parameter
    ZSTD_CCtx_setParameter(cctx, ZSTD_c_nbWorkers, 3);
    ZSTD_CCtx_setParameter(cctx, ZSTD_c_compressionLevel, ZSTD_btlazy2);

    com_size = ZSTD_compress2(cctx, com_ptr, com_space_size, file_text_ptr, file_len);

    free(com_ptr);
    free(file_text_ptr);
    return 0;
}

執行上述demo，可見zstd確實啟動了3個執行緒對文字進行了並行壓縮。且設定的執行緒數越多，壓縮時間越短，這裡就不詳細展示了，讀者可以自行實驗。

需要說明的是，zstd當前預設編譯單執行緒的庫檔案，要實現多執行緒的API呼叫，需要在make的時候指定編譯引數ZSTD_MULTITHREAD。