Minor fix (#47)

* add SIEVE to be the main cache of TinyLFU algorithm * log fix * Add options for trace print formatting * add a note how to run the caffeine simulator * Update table of contents in README.md
1a1a11a · Jan 4, 2024 · 938165f · 938165f
1 parent 4d47e20
commit 938165f
Show file tree

Hide file tree

Showing 7 changed files with 86 additions and 32 deletions.
diff --git a/README.md b/README.md
@@ -4,22 +4,30 @@
 
 
 <!-- TOC start (generated with https://github.com/derlin/bitdowntoc) -->
-* [What is libCacheSim](#what-is-libcachesim)
-* [libCacheSim features ](#libcachesim-features)
-* [Supported algorithms](#supported-algorithms)
-* [Build and Install libCacheSim](#build-and-install-libcachesim)
-  + [One-line install](#one-line-install)
-  + [Install dependency](#install-dependency)
-  + [Build libCacheSim](#build-libcachesim)
-* [Usage](#usage)
-  + [cachesim (a high-performance cache simulator)](#cachesim-a-high-performance-cache-simulator)
-  + [Trace analysis](#trace-analysis)
-  + [Using libCacheSim as a library ](#using-libcachesim-as-a-library)
-  + [Extending libCacheSim (new algorithms and trace types)](#extending-libcachesim-new-algorithms-and-trace-types)
-* [Open source cache traces](#open-source-cache-traces)
-* [Questions? ](#questions)
-* [Reference](#reference)
-* [License](#license)
+- [libCacheSim - building and running cache simulations](#libcachesim---building-and-running-cache-simulations)
+  - [News](#news)
+  - [What is libCacheSim](#what-is-libcachesim)
+  - [libCacheSim features](#libcachesim-features)
+  - [Supported algorithms](#supported-algorithms)
+  - [Build and Install libCacheSim](#build-and-install-libcachesim)
+    - [One-line install](#one-line-install)
+    - [Install dependency](#install-dependency)
+    - [Build libCacheSim](#build-libcachesim)
+  - [Usage](#usage)
+    - [cachesim (a high-performance cache simulator)](#cachesim-a-high-performance-cache-simulator)
+      - [basic usage](#basic-usage)
+      - [Run a single cache simulation](#run-a-single-cache-simulation)
+      - [Run multiple cache simulations with different cache sizes](#run-multiple-cache-simulations-with-different-cache-sizes)
+      - [Plot miss ratio curve](#plot-miss-ratio-curve)
+    - [Trace analysis](#trace-analysis)
+    - [Using libCacheSim as a library](#using-libcachesim-as-a-library)
+    - [Extending libCacheSim (new algorithms and trace types)](#extending-libcachesim-new-algorithms-and-trace-types)
+  - [Open source cache traces](#open-source-cache-traces)
+  - [Questions?](#questions)
+  - [Contributions](#contributions)
+  - [Reference](#reference)
+  - [License](#license)
+  - [Related](#related)
 <!-- TOC end -->
 
 
@@ -251,7 +259,7 @@ The compressed traces can be used with libCacheSim without decompression. And li
 | Tencent Photo | 2018 |   object  |                      [link](http://iotta.snia.org/traces/parallel?only=27476)                     | [link](https://ftp.pdl.cmu.edu/pub/datasets/twemcacheWorkload/cacheDatasets/tencentPhoto/) |
 | WikiCDN       | 2019 |   object  |          [link](https://wikitech.wikimedia.org/wiki/Analytics/Data_Lake/Traffic/Caching)          |     [link](https://ftp.pdl.cmu.edu/pub/datasets/twemcacheWorkload/cacheDatasets/wiki/)     |
 | Tencent CBS   | 2020 |   block   |                      [link](http://iotta.snia.org/traces/parallel?only=27917)                     | [link](https://ftp.pdl.cmu.edu/pub/datasets/twemcacheWorkload/cacheDatasets/tencentBlock/) |
-| Alibaba CBS   | 2020 |   block   |                          [link](https://github.com/alibaba/block-traces)                          | [link](https://ftp.pdl.cmu.edu/pub/datasets/twemcacheWorkload/cacheDatasets/alibabaBlock/) |
+| Alibaba Block | 2020 |   block   |                          [link](https://github.com/alibaba/block-traces)                          | [link](https://ftp.pdl.cmu.edu/pub/datasets/twemcacheWorkload/cacheDatasets/alibabaBlock/) |
 | Twitter       | 2020 | key-value |                          [link](https://github.com/twitter/cache-traces)                          | [link](https://ftp.pdl.cmu.edu/pub/datasets/twemcacheWorkload/cacheDatasets/twitter/)      |
 | MetaKV        | 2022 | key-value | [link](https://cachelib.org/docs/Cache_Library_User_Guides/Cachebench_FB_HW_eval/#list-of-traces) | [link](https://ftp.pdl.cmu.edu/pub/datasets/twemcacheWorkload/cacheDatasets/metaKV/)       |
 | MetaCDN       | 2023 | object    | [link](https://cachelib.org/docs/Cache_Library_User_Guides/Cachebench_FB_HW_eval/#list-of-traces) | [link](https://ftp.pdl.cmu.edu/pub/datasets/twemcacheWorkload/cacheDatasets/metaCDN/)      |

diff --git a/libCacheSim/bin/traceUtils/cli_parser.cpp b/libCacheSim/bin/traceUtils/cli_parser.cpp
@@ -30,6 +30,9 @@ enum argp_option_short {
 
   // trace print
   OPTION_NUM_REQ = 'n',
+  OPTION_FIELD_DELIMITER = 0x201,
+  OPTION_OBJ_ID_ONLY = 0x202,
+  OPTION_OBJ_ID_32bit = 0x204,
 
   // trace filter
   OPTION_FILTER_TYPE = 0x301,
@@ -63,6 +66,11 @@ static struct argp_option options[] = {
     {0, 0, 0, 0, "tracePrint options:"},
     {"num-req", OPTION_NUM_REQ, "-1", 0,
      "Number of requests to process, -1 means all requests in the trace", 6},
+    {"field-delimiter", OPTION_FIELD_DELIMITER, ",", 0,
+     "The delimiter formatting the trace", 6},
+    {"obj-id-only", OPTION_OBJ_ID_ONLY, "0", 0, "Only to print object id", 6},
+    {"obj-id-32bit", OPTION_OBJ_ID_32bit, "0", 0,
+     "Print object id as 32-bit int", 6},
 
     {0, 0, 0, 0, "traceFilter options:"},
     {"filter-type", OPTION_FILTER_TYPE, "FIFO", 0,
@@ -106,6 +114,15 @@ static error_t parse_opt(int key, char *arg, struct argp_state *state) {
     case OPTION_NUM_REQ:
       arguments->n_req = atoll(arg);
       break;
+    case OPTION_FIELD_DELIMITER:
+      arguments->delimiter = arg[0];
+      break;
+    case OPTION_OBJ_ID_ONLY:
+      arguments->print_obj_id_only = atol(arg);
+      break;
+    case OPTION_OBJ_ID_32bit:
+      arguments->print_obj_id_32bit = atol(arg);
+      break;
     case OPTION_FILTER_TYPE:
       arguments->cache_name = arg;
       break;
@@ -146,6 +163,8 @@ static char doc[] =
     "tracePrint: utility to print binary trace in human-readable format\n"
     "traceConv: utility to convert a trace to oracleGeneral format\n\n"
     "traceFilter: utility to filter a trace\n\n"
+    "example usage: ./tracePrint /trace/path oracleGeneral -n 20 "
+    "--obj-id-only=1\n\n"
     "example usage: ./traceConv /trace/path csv -o "
     "/path/new_trace.oracleGeneral -t "
     "\"obj-id-col=5,time-col=2,obj-size-col=4\"\n\n"
@@ -171,6 +190,9 @@ static void init_arg(struct arguments *args) {
   args->remove_size_change = false;
   args->cache_name = NULL;
   args->cache_size = 0;
+  args->delimiter = ',';
+  args->print_obj_id_only = false;
+  args->print_obj_id_32bit = false;
 }
 
 static void print_parsed_arg(struct arguments *args) {
@@ -229,7 +251,8 @@ void parse_cmd(int argc, char *argv[], struct arguments *args) {
   assert(N_ARGS == 2);
 
   args->reader = create_reader(args->trace_type_str, args->trace_path,
-                               args->trace_type_params, args->n_req, args->ignore_obj_size, 0);
+                               args->trace_type_params, args->n_req,
+                               args->ignore_obj_size, 0);
 
   print_parsed_arg(args);
 }

diff --git a/libCacheSim/bin/traceUtils/internal.hpp b/libCacheSim/bin/traceUtils/internal.hpp
@@ -31,6 +31,9 @@ struct arguments {
 
   /* trace print */
   int64_t num_req; /* number of requests to print */
+  char delimiter;
+  bool print_obj_id_only;
+  bool print_obj_id_32bit;
 
   /* trace filter */
   char *cache_name;

diff --git a/libCacheSim/bin/traceUtils/tracePrintMain.cpp b/libCacheSim/bin/traceUtils/tracePrintMain.cpp
@@ -8,30 +8,39 @@
 #include "../../include/libCacheSim/reader.h"
 #include "internal.hpp"
 
-
 int main(int argc, char *argv[]) {
   struct arguments args;
 
   cli::parse_cmd(argc, argv, &args);
 
   request_t *req = new_request();
   read_one_req(args.reader, req);
-  
+
   bool trace_has_next_access_vtime = req->next_access_vtime != -2;
 
-  if (trace_has_next_access_vtime) {
-    printf("# time, object, size, next_access_vtime\n");
-  } else {
-    printf("# time, object, size\n");
+  if (!args.print_obj_id_only) {
+    if (trace_has_next_access_vtime) {
+      printf("# time,object,size,next_access_vtime\n");
+    } else {
+      printf("# time,object,size\n");
+    }
   }
 
   while (req->valid) {
-    printf("%ld, %lu, %d", (long)req->clock_time, (unsigned long)req->obj_id,
-           (int)req->obj_size);
-    if (trace_has_next_access_vtime) {
-      printf(", %ld\n", (long)req->next_access_vtime);
+    if (args.print_obj_id_32bit) {
+      req->obj_id = (uint32_t)req->obj_id;
+    }
+
+    if (args.print_obj_id_only) {
+      printf("%lu\n", (unsigned long)req->obj_id);
     } else {
-      printf("\n");
+      printf("%ld%c%lu%c%d", (long)req->clock_time, args.delimiter,
+             (unsigned long)req->obj_id, args.delimiter, (int)req->obj_size);
+      if (trace_has_next_access_vtime) {
+        printf("%c%ld\n", args.delimiter, (long)req->next_access_vtime);
+      } else {
+        printf("\n");
+      }
     }
     read_one_req(args.reader, req);
   }
@@ -40,4 +49,3 @@ int main(int argc, char *argv[]) {
 
   return 0;
 }
-
diff --git a/libCacheSim/cache/eviction/WTinyLFU.c b/libCacheSim/cache/eviction/WTinyLFU.c
@@ -138,6 +138,8 @@ cache_t *WTinyLFU_init(const common_cache_params_t ccache_params,
     params->main_cache = Hyperbolic_init(ccache_params_local, NULL);
   } else if (strcasecmp(params->main_cache_type, "LHD") == 0) {
     params->main_cache = LHD_init(ccache_params_local, NULL);
+  } else if (strcasecmp(params->main_cache_type, "SIEVE") == 0) {
+    params->main_cache = Sieve_init(ccache_params_local, NULL);
   } else {
     ERROR("WTinyLFU does not support %s \n", params->main_cache_type);
   }

diff --git a/libCacheSim/traceAnalyzer/analyzer.cpp b/libCacheSim/traceAnalyzer/analyzer.cpp
@@ -110,8 +110,9 @@ void traceAnalyzer::TraceAnalyzer::run() {
     if (curr_time_window_idx != time_to_window_idx(req->clock_time)) {
       ERROR(
           "The data is not ordered by time, please sort the trace first!"
-          "Current time %ld requested object %lu\n",
-          (long) req->clock_time + start_ts_, (unsigned long) req->obj_id);
+          "Current time %ld requested object %lu, obj size %lu\n",
+          (long)req->clock_time + start_ts_, (unsigned long)req->obj_id,
+          req->obj_size);
     }
 
     DEBUG_ASSERT(curr_time_window_idx == time_to_window_idx(req->clock_time));

diff --git a/scripts/note b/scripts/note
@@ -0,0 +1,9 @@
+## How to run Caffeine simulator
+```
+git clone https://github.com/ben-manes/caffeine.git
+./gradlew build
+export GRADLE_OPTS="-Xmx204800m"
+# modify simulator/src/main/resources/reference.conf, the lirs format is txt with only the object id
+./gradlew run simulator:run
+
+```