Update server configs generating scripts

scripts/load_testing/README.md

@@ -0,0 +1,30 @@
+## Examples
+
+For PMEM usage use `-m` parameter followed by a list of PMEM mount point paths. Without that parameter configs will be created for RAM usage.
+Note that the first mount point provided will be bound to the numa node 0, next mount point to the numa node 1, etc.
+Put a list of mount points in quotes when providing more than one path.
+
+Twemcache:
+
+```
+# generating configs without PMEM support
+./generate.sh -s -p pelikan/_build/_bin/pelikan_twemcache -c -r rpc-perf/target/release/rpc-perf -t 127.0.0.1
+
+# generating configs with PMEM support for two mount points
+./generate.sh -s -p pelikan/_build/_bin/pelikan_twemcache -c -r rpc-perf/target/release/rpc-perf -t 127.0.0.1 -m "/mnt/pmem0 /mnt/pmem1"
+```
+
+Slimcache:
+
+```
+# generating configs without PMEM support
+./generate.sh -s -p pelikan/_build/_bin/pelikan_slimcache -c -r rpc-perf/target/release/rpc-perf -t 127.0.0.1
+
+# generating configs with PMEM support for two mount points
+./generate.sh -s -p pelikan/_build/_bin/pelikan_slimcache -c -r rpc-perf/target/release/rpc-perf -t 127.0.0.1 -m "/mnt/pmem0 /mnt/pmem1"
+```
+
+To run benchmarks provide paths to generated config directories:
+```
+./runtest.sh -c rpcperf_100_1024_4 -s pelikan_1024_4 -t 127.0.0.1
+```

scripts/load_testing/client_config.py

@@ -60,7 +60,7 @@ def generate_runscript(binary, server_ip, instances):
       the_file.write(' --waterfall latency-waterfall-{server_port}.png'.format(server_port=server_port))
       the_file.write(' > rpcperf_{server_port}.log'.format(server_port=server_port))
       the_file.write(' 2>&1 &\n')
-  os.chmod(fname, 0777)
+  os.chmod(fname, 0o777)
 
 
 if __name__ == "__main__":

scripts/load_testing/generate.sh

@@ -14,15 +14,17 @@ server=false
 rpcperf="rpc-perf"
 pelikan="pelikan_twemcache"
 target="127.0.0.1"
+pmem_paths=()
 
 show_help()
 {
-    echo "generate.sh [-c [-r path/to/rpcperf] [-t target/serverIP]] [-s [-p path/to/pelikan]]"
+    echo 'generate.sh [-c [-r path/to/rpcperf] [-t target/serverIP]] [-s [-p path/to/pelikan]] [-m "path/to/pmem0 path/to/pmem1 ..."]'
+    echo 'Note that the first pmem path is bound to the first numa node, the second path is bound to the next numa node. One or more paths can be provided.'
 }
 
 get_args()
 {
-    while getopts ":p:r:t:csh" opt; do
+    while getopts ":p:r:t:m:csh" opt; do
         case "$opt" in
         c)  client=true
             ;;
@@ -34,6 +36,8 @@ get_args()
             ;;
         t)  target=$OPTARG
             ;;
+        m)  pmem_paths=($OPTARG)
+            ;;
         h)
             show_help
             exit 0
@@ -57,7 +61,7 @@ gen_pelikan()
         do
             slab_mem=$((mem * 1024 * 1024 * 1024))
             prefix=pelikan_${size}_${mem}
-            python server_config.py --prefix="$prefix" --binary="$pelikan" --instances="$instances" --slab_mem "$slab_mem" --vsize "$vsize"
+            python server_config.py --prefix="$prefix" --binary="$pelikan" --instances="$instances" --slab_mem "$slab_mem" --vsize "$vsize" --pmem_paths ${pmem_paths[@]}
         done
     done
 }
@@ -87,4 +91,3 @@ fi
 if [ "$server" = true ]; then
     gen_pelikan
 fi
-

scripts/load_testing/server_config.py

@@ -1,6 +1,8 @@
 import argparse
 from math import ceil, floor, log
 import os
+import subprocess
+import sys
 
 INSTANCES = 3
 PREFIX = 'test'
@@ -13,8 +15,9 @@
 THREAD_PER_SOCKET = 48
 BIND_TO_CORES = False
 BIND_TO_NODES = True
+ENGINE = "twemcache"
 
-def generate_config(instances, vsize, slab_mem):
+def generate_config(instances, vsize, slab_mem, pmem_paths, engine):
   # create top-level folders under prefix
   try:
     os.makedirs('config')
@@ -25,14 +28,17 @@ def generate_config(instances, vsize, slab_mem):
   except:
     pass
 
-  nkey = int(ceil(1.0 * slab_mem / (vsize + KSIZE + PELIKAN_ITEM_OVERHEAD)))
+  item_size = vsize + KSIZE + PELIKAN_ITEM_OVERHEAD
+  nkey = int(ceil(1.0 * slab_mem / item_size))
   hash_power = int(ceil(log(nkey, 2)))
 
-  # create twemcache config file(s)
+  # create twemcache|slimcache config file(s)
   for i in range(instances):
     admin_port = PELIKAN_ADMIN_PORT + i
     server_port = PELIKAN_SERVER_PORT + i
-    config_file = 'twemcache-{server_port}.config'.format(server_port=server_port)
+    config_file = '{engine}-{server_port}.config'.format(engine=engine, server_port=server_port)
+
+    # String with common options for both twemcache and slimcache
     config_str = """\
 daemonize: yes
 admin_port: {admin_port}
@@ -45,11 +51,11 @@ def generate_config(instances, vsize, slab_mem):
 buf_sock_poolsize: 16384
 
 debug_log_level: 5
-debug_log_file: log/twemcache-{server_port}.log
+debug_log_file: log/{engine}-{server_port}.log
 debug_log_nbuf: 1048576
 
-klog_file: log/twemcache-{server_port}.cmd
-klog_backup: log/twemcache-{server_port}.cmd.old
+klog_file: log/{engine}-{server_port}.cmd
+klog_backup: log/{engine}-{server_port}.cmd.old
 klog_sample: 100
 klog_max: 1073741824
 
@@ -61,70 +67,109 @@ def generate_config(instances, vsize, slab_mem):
 request_poolsize: 16384
 response_poolsize: 32768
 
+time_type: 2
+""".format(admin_port=admin_port, server_port=server_port, vsize=vsize, nkey=nkey, engine=engine)
+
+    # String with options specific for either twemcache or slimcache
+    pmem_path_str = ""
+    datapool_param = ""
+    if engine == "slimcache":
+      datapool_param = "cuckoo_datapool"
+      engine_str = """\
+
+cuckoo_item_size: {item_size}
+cuckoo_nitem: {nkey}
+cuckoo_datapool_prefault: yes
+""".format(item_size=item_size, nkey=nkey)
+    elif engine == "twemcache":
+      datapool_param = "slab_datapool"
+      engine_str = """\
+
 slab_evict_opt: 1
 slab_prealloc: yes
 slab_hash_power: {hash_power}
 slab_mem: {slab_mem}
 slab_size: 1048756
+slab_datapool_prefault: yes
 
 stats_intvl: 10000
 stats_log_file: log/twemcache-{server_port}.stats
+""".format(hash_power=hash_power, slab_mem=slab_mem, server_port=server_port)
 
-time_type: 2
-""".format(admin_port=admin_port, server_port=server_port, vsize=vsize, nkey=nkey, hash_power=hash_power, slab_mem=slab_mem)
+    # String with option specific for PMEM usage
+    if len(pmem_paths) > 0:
+        pmem_path_str = """\
+
+{datapool_param}: {pmem_path}
+""".format(datapool_param=datapool_param, pmem_path=os.path.join(pmem_paths[i%len(pmem_paths)], 'pool_{}'.format(server_port)))
+
+    # Put it all together
+    config_str = config_str + engine_str + pmem_path_str
     with open(os.path.join('config', config_file),'w') as the_file:
       the_file.write(config_str)
 
-def generate_runscript(binary, instances):
+def generate_runscript(binary, instances, pmem_paths_count, engine):
   # create bring-up.sh
   fname = 'bring-up.sh'
+  numa_node_count = pmem_paths_count if pmem_paths_count > 0 else 2
   with open(fname, 'w') as the_file:
     for i in range(instances):
-      config_file = os.path.join('config', 'twemcache-{server_port}.config'.format(server_port=PELIKAN_SERVER_PORT+i))
+      config_file = os.path.join('config', '{engine}-{server_port}.config'.format(engine=engine, server_port=PELIKAN_SERVER_PORT+i))
       if BIND_TO_NODES:
         the_file.write('sudo numactl --cpunodebind={numa_node} --preferred={numa_node} '.format(
-            numa_node=i%2))
+            numa_node=i%numa_node_count))
       elif BIND_TO_CORES:
         the_file.write('sudo numactl --physcpubind={physical_thread},{logical_thread} '.format(
             physical_thread=i,
             logical_thread=i+THREAD_PER_SOCKET))
       the_file.write('{binary_file} {config_file}\n'.format(
           binary_file=binary,
           config_file=config_file))
-  os.chmod(fname, 0777)
+  os.chmod(fname, 0o777)
 
   # create warm-up.sh
   fname = 'warm-up.sh'
+  prefill_opt = "prefilling cuckoo" if engine == "slimcache" else "prefilling slab"
   with open(fname, 'w') as the_file:
     the_file.write("""
 ./bring-up.sh
 
 nready=0
 while [ $nready -lt {instances} ]
 do
-    nready=$(grep -l "prefilling slab" log/twemcache-*.log | wc -l)
+    nready=$(grep -l "{prefill_opt}" log/{engine}-*.log | wc -l)
     echo "$(date): $nready out of {instances} servers are warmed up"
     sleep 10
 done
-""".format(instances=instances))
-  os.chmod(fname, 0777)
+""".format(instances=instances, prefill_opt=prefill_opt, engine=engine))
+  os.chmod(fname, 0o777)
 
 
 if __name__ == "__main__":
   parser = argparse.ArgumentParser(description="""
     Generate all the server-side scripts/configs needed for a test run.
     """)
-  parser.add_argument('--binary', dest='binary', type=str, help='location of pelikan_twemcache binary', required=True)
+  parser.add_argument('--binary', dest='binary', type=str, help='location of pelikan_twemcache|pelikan_slimcache binary', required=True)
   parser.add_argument('--prefix', dest='prefix', type=str, default=PREFIX, help='folder that contains all the other files to be generated')
   parser.add_argument('--instances', dest='instances', type=int, default=INSTANCES, help='number of instances')
   parser.add_argument('--vsize', dest='vsize', type=int, default=VSIZE, help='value size')
   parser.add_argument('--slab_mem', dest='slab_mem', type=int, default=PELIKAN_SLAB_MEM, help='total capacity of slab memory, in bytes')
+  parser.add_argument('--pmem_paths', dest='pmem_paths', nargs='*', help='list of pmem mount points')
 
   args = parser.parse_args()
 
   if not os.path.exists(args.prefix):
     os.makedirs(args.prefix)
   os.chdir(args.prefix)
 
-  generate_config(args.instances, args.vsize, args.slab_mem)
-  generate_runscript(args.binary, args.instances)
+  binary_help_out = subprocess.check_output([args.binary, '--help'])
+  if binary_help_out.find("twemcache") != -1:
+    engine = "twemcache"
+  elif binary_help_out.find("slimcache") != -1:
+    engine = "slimcache"
+  else:
+    print('Provided binary is not twemcache|slimcache. Only these engines are valid. Exiting...')
+    sys.exit()
+
+  generate_config(args.instances, args.vsize, args.slab_mem, args.pmem_paths, engine)
+  generate_runscript(args.binary, args.instances, len(args.pmem_paths), engine)