Experiment/WIP: Try to track "secondary" recovery ops separately

src/messenger.h

@@ -149,7 +149,7 @@ public:
     std::map<osd_num_t, osd_wanted_peer_t> wanted_peers;
     std::map<uint64_t, int> osd_peer_fds;
     // op statistics
-    osd_op_stats_t stats;
+    osd_op_stats_t stats, recovery_stats;
 
     void init();
     void parse_config(const json11::Json & config);
@@ -175,6 +175,7 @@ public:
     bool connect_rdma(int peer_fd, std::string rdma_address, uint64_t client_max_msg);
 #endif
 
+    void inc_op_stats(osd_op_stats_t & stats, uint64_t opcode, timespec & tv_begin, timespec & tv_end, uint64_t len);
     void measure_exec(osd_op_t *cur_op);
 
 protected:

src/msgr_op.cpp

@@ -24,3 +24,17 @@ osd_op_t::~osd_op_t()
         free(buf);
     }
 }
+
+bool osd_op_t::is_recovery_related()
+{
+    return (req.hdr.opcode == OSD_OP_SEC_READ ||
+        req.hdr.opcode == OSD_OP_SEC_WRITE ||
+        req.hdr.opcode == OSD_OP_SEC_WRITE_STABLE) &&
+        (req.sec_rw.flags & OSD_OP_RECOVERY_RELATED) ||
+        req.hdr.opcode == OSD_OP_SEC_DELETE &&
+        (req.sec_del.flags & OSD_OP_RECOVERY_RELATED) ||
+        req.hdr.opcode == OSD_OP_SEC_STABILIZE &&
+        (req.sec_stab.flags & OSD_OP_RECOVERY_RELATED) ||
+        req.hdr.opcode == OSD_OP_SEC_SYNC &&
+        (req.sec_sync.flags & OSD_OP_RECOVERY_RELATED);
+}

src/msgr_op.h

@@ -173,4 +173,6 @@ struct osd_op_t
     osd_op_buf_list_t iov;
 
     ~osd_op_t();
+
+    bool is_recovery_related();
 };

src/msgr_send.cpp

@@ -131,6 +131,23 @@ void osd_messenger_t::outbox_push(osd_op_t *cur_op)
     }
 }
 
+void osd_messenger_t::inc_op_stats(osd_op_stats_t & stats, uint64_t opcode, timespec & tv_begin, timespec & tv_end, uint64_t len)
+{
+    uint64_t usecs = (
+        (tv_end.tv_sec - tv_begin.tv_sec)*1000000 +
+        (tv_end.tv_nsec - tv_begin.tv_nsec)/1000
+    );
+    stats.op_stat_count[opcode]++;
+    if (!stats.op_stat_count[opcode])
+    {
+        stats.op_stat_count[opcode] = 1;
+        stats.op_stat_sum[opcode] = 0;
+        stats.op_stat_bytes[opcode] = 0;
+    }
+    stats.op_stat_sum[opcode] += usecs;
+    stats.op_stat_bytes[opcode] += len;
+}
+
 void osd_messenger_t::measure_exec(osd_op_t *cur_op)
 {
     // Measure execution latency
@@ -142,29 +159,24 @@ void osd_messenger_t::measure_exec(osd_op_t *cur_op)
     {
         clock_gettime(CLOCK_REALTIME, &cur_op->tv_end);
     }
-    stats.op_stat_count[cur_op->req.hdr.opcode]++;
+    uint64_t len = 0;
-    if (!stats.op_stat_count[cur_op->req.hdr.opcode])
-    {
-        stats.op_stat_count[cur_op->req.hdr.opcode]++;
-        stats.op_stat_sum[cur_op->req.hdr.opcode] = 0;
-        stats.op_stat_bytes[cur_op->req.hdr.opcode] = 0;
-    }
-    stats.op_stat_sum[cur_op->req.hdr.opcode] += (
-        (cur_op->tv_end.tv_sec - cur_op->tv_begin.tv_sec)*1000000 +
-        (cur_op->tv_end.tv_nsec - cur_op->tv_begin.tv_nsec)/1000
-    );
     if (cur_op->req.hdr.opcode == OSD_OP_READ ||
         cur_op->req.hdr.opcode == OSD_OP_WRITE ||
         cur_op->req.hdr.opcode == OSD_OP_SCRUB)
     {
         // req.rw.len is internally set to the full object size for scrubs
-        stats.op_stat_bytes[cur_op->req.hdr.opcode] += cur_op->req.rw.len;
+        len = cur_op->req.rw.len;
     }
     else if (cur_op->req.hdr.opcode == OSD_OP_SEC_READ ||
         cur_op->req.hdr.opcode == OSD_OP_SEC_WRITE ||
         cur_op->req.hdr.opcode == OSD_OP_SEC_WRITE_STABLE)
     {
-        stats.op_stat_bytes[cur_op->req.hdr.opcode] += cur_op->req.sec_rw.len;
+        len = cur_op->req.sec_rw.len;
+    }
+    inc_op_stats(stats, cur_op->req.hdr.opcode, cur_op->tv_begin, cur_op->tv_end, len);
+    if (cur_op->is_recovery_related())
+    {
+        inc_op_stats(recovery_stats, cur_op->req.hdr.opcode, cur_op->tv_begin, cur_op->tv_end, len);
     }
 }
 

src/osd.h

@@ -123,7 +123,7 @@ class osd_t
     double recovery_tune_max_util = 1.0;
     double recovery_tune_max_client_util = 0.5;
     int recovery_tune_interval = 1;
-    double recovery_tune_ewma_rate = 0.5;
+    double recovery_tune_ewma_rate = 0.2;
     int recovery_tune_sleep_min_us = 10;
     int recovery_pg_switch = DEFAULT_RECOVERY_PG_SWITCH;
     int recovery_sync_batch = DEFAULT_RECOVERY_BATCH;
@@ -208,10 +208,8 @@ class osd_t
     // recovery auto-tuning
     int rtune_timer_id = -1;
     uint64_t rtune_avg_lat = 0;
-    double rtune_avg_count = 0;
     double rtune_client_util = 0, rtune_target_util = 1;
-    osd_op_stats_t rtune_prev_stats;
+    osd_op_stats_t rtune_prev_stats, rtune_prev_recovery_stats;
-    recovery_stat_t rtune_prev_recovery[2];
     uint64_t recovery_target_queue_depth = 1;
     uint64_t recovery_target_sleep_us = 0;
 
@@ -304,7 +302,7 @@ class osd_t
     bool remember_unstable_write(osd_op_t *cur_op, pg_t & pg, pg_osd_set_t & loc_set, int base_state);
     void handle_primary_subop(osd_op_t *subop, osd_op_t *cur_op);
     void handle_primary_bs_subop(osd_op_t *subop);
-    void add_bs_subop_stats(osd_op_t *subop);
+    void add_bs_subop_stats(osd_op_t *subop, bool recovery_related = false);
     void pg_cancel_write_queue(pg_t & pg, osd_op_t *first_op, object_id oid, int retval);
 
     void submit_primary_subops(int submit_type, uint64_t op_version, const uint64_t* osd_set, osd_op_t *cur_op);

src/osd_flush.cpp

@@ -325,17 +325,7 @@ void osd_t::submit_recovery_op(osd_recovery_op_t *op)
         {
             printf("Recovery operation done for %lx:%lx\n", op->oid.inode, op->oid.stripe);
         }
-        if (recovery_target_sleep_us)
+        finish_recovery_op(op);
-        {
-            this->tfd->set_timer_us(recovery_target_sleep_us, false, [this, op](int timer_id)
-            {
-                finish_recovery_op(op);
-            });
-        }
-        else
-        {
-            finish_recovery_op(op);
-        }
     };
     exec_op(op->osd_op);
 }
@@ -383,29 +373,46 @@ void osd_t::finish_recovery_op(osd_recovery_op_t *op)
 
 void osd_t::tune_recovery()
 {
-    static int total_client_ops[] = { OSD_OP_READ, OSD_OP_WRITE, OSD_OP_SYNC, OSD_OP_DELETE };
+    static int accounted_ops[] = {
-    uint64_t total_client_usec = 0;
+        OSD_OP_SEC_READ, OSD_OP_SEC_WRITE, OSD_OP_SEC_WRITE_STABLE,
-    for (int i = 0; i < sizeof(total_client_ops)/sizeof(total_client_ops[0]); i++)
+        OSD_OP_SEC_STABILIZE, OSD_OP_SEC_SYNC, OSD_OP_SEC_DELETE
+    };
+    uint64_t total_client_usec = 0, total_recovery_usec = 0, recovery_count = 0;
+    for (int i = 0; i < sizeof(accounted_ops)/sizeof(accounted_ops[0]); i++)
     {
-        total_client_usec += (msgr.stats.op_stat_sum[total_client_ops[i]] - rtune_prev_stats.op_stat_sum[total_client_ops[i]]);
+        total_client_usec += (msgr.stats.op_stat_sum[accounted_ops[i]]
-        rtune_prev_stats.op_stat_sum[total_client_ops[i]] = msgr.stats.op_stat_sum[total_client_ops[i]];
+            - rtune_prev_stats.op_stat_sum[accounted_ops[i]]);
+        total_recovery_usec += (msgr.recovery_stats.op_stat_sum[accounted_ops[i]]
+            - rtune_prev_recovery_stats.op_stat_sum[accounted_ops[i]]);
+        recovery_count += (msgr.recovery_stats.op_stat_count[accounted_ops[i]]
+            - rtune_prev_recovery_stats.op_stat_count[accounted_ops[i]]);
+        rtune_prev_stats.op_stat_sum[accounted_ops[i]] = msgr.stats.op_stat_sum[accounted_ops[i]];
+        rtune_prev_recovery_stats.op_stat_sum[accounted_ops[i]] = msgr.recovery_stats.op_stat_sum[accounted_ops[i]];
+        rtune_prev_recovery_stats.op_stat_count[accounted_ops[i]] = msgr.recovery_stats.op_stat_count[accounted_ops[i]];
     }
-    uint64_t total_recovery_usec = 0, recovery_count = 0;
+    total_client_usec -= total_recovery_usec;
-    total_recovery_usec += recovery_stat[0].usec-rtune_prev_recovery[0].usec;
-    total_recovery_usec += recovery_stat[1].usec-rtune_prev_recovery[1].usec;
-    recovery_count += recovery_stat[0].count-rtune_prev_recovery[0].count;
-    recovery_count += recovery_stat[1].count-rtune_prev_recovery[1].count;
-    memcpy(rtune_prev_recovery, recovery_stat, sizeof(recovery_stat));
     if (recovery_count == 0)
     {
         return;
     }
-    rtune_avg_lat = total_recovery_usec/recovery_count*recovery_tune_ewma_rate +
+    // example:
-        rtune_avg_lat*(1-recovery_tune_ewma_rate);
+    // total 3 GB/s
-    rtune_avg_count = recovery_count*recovery_tune_ewma_rate +
+    // recovery queue 1
-        rtune_avg_count*(1-recovery_tune_ewma_rate);
+    // 120 OSDs
-    // client_util = count/interval * usec/1000000.0/count = usec/1000000.0/interval :-)
+    // EC 5+3
-    double client_util = total_client_usec/1000000.0/recovery_tune_interval;
+    // 128kb block_size => 640kb object
+    // 3000*1024/640/120 = 40 MB/s per OSD = 64 recovered objects per OSD
+    //   = 64*8*2 subops = 1024 recovery subop iops
+    // 8 recovery subop queue
+    // => subop avg latency = 0.0078125 sec
+    // utilisation = 8
+    // target util 1
+    // intuitively target latency should be 8x of real
+    // target_lat = rtune_avg_lat * utilisation / target_util
+    //            = rtune_avg_lat * rtune_avg_lat * rtune_avg_iops / target_util
+    //            = 0.0625
+    // recovery utilisation will be 1
+    auto client_util = total_client_usec/1000000.0/recovery_tune_interval;
     rtune_client_util = rtune_client_util*(1-recovery_tune_ewma_rate) + client_util*recovery_tune_ewma_rate;
     rtune_target_util = (rtune_client_util < recovery_tune_min_client_util
         ? recovery_tune_max_util
@@ -414,19 +421,15 @@ void osd_t::tune_recovery()
                 (recovery_tune_max_client_util-rtune_client_util)/(recovery_tune_max_client_util-recovery_tune_min_client_util)
         )
     );
-    // for example: utilisation = 8, target = 1
+    rtune_avg_lat = total_recovery_usec/recovery_count*recovery_tune_ewma_rate + rtune_avg_lat*(1-recovery_tune_ewma_rate);
-    // intuitively target latency should be 8x of real
-    // target_lat = rtune_avg_lat * utilisation / target_util
-    //            = rtune_avg_lat * rtune_avg_lat * rtune_avg_iops / target_util
-    //            = 0.0625
     recovery_target_queue_depth = (int)rtune_target_util + (rtune_target_util < 1 || rtune_target_util-(int)rtune_target_util >= 0.1 ? 1 : 0);
-    uint64_t target_lat = rtune_avg_lat * rtune_avg_lat/1000000.0*rtune_avg_count/recovery_tune_interval/rtune_target_util;
+    uint64_t target_lat = rtune_avg_lat * rtune_avg_lat/1000000.0 * recovery_count/recovery_tune_interval / rtune_target_util;
     recovery_target_sleep_us = target_lat > rtune_avg_lat+recovery_tune_sleep_min_us ? target_lat-rtune_avg_lat : 0;
     if (log_level > 3)
     {
         printf(
-            "recovery tune: client util %.2f (ewma %.2f), target util %.2f -> queue %ld, lat %lu us, real %lu us, pause %lu us\n",
+            "recovery tune: cli %lu us, recovery %lu us / %lu ops, target util %.2f -> queue %ld, lat %lu us, real %lu us, delay %lu us\n",
-            client_util, rtune_client_util, rtune_target_util, recovery_target_queue_depth, target_lat, rtune_avg_lat, recovery_target_sleep_us
+            total_client_usec, total_recovery_usec, recovery_count, rtune_target_util, recovery_target_queue_depth, target_lat, rtune_avg_lat, recovery_target_sleep_us
         );
     }
 }

src/osd_ops.h

@@ -34,6 +34,7 @@
 #define OSD_OP_MAX                  18
 #define OSD_RW_MAX                  64*1024*1024
 #define OSD_PROTOCOL_VERSION        1
+#define OSD_OP_RECOVERY_RELATED     (uint32_t)1
 
 // Memory alignment for direct I/O (usually 512 bytes)
 #ifndef DIRECT_IO_ALIGNMENT
@@ -88,7 +89,8 @@ struct __attribute__((__packed__)) osd_op_sec_rw_t
     uint32_t len;
     // bitmap/attribute length - bitmap comes after header, but before data
     uint32_t attr_len;
-    uint32_t pad0;
+    // the only possible flag is OSD_OP_RECOVERY_RELATED
+    uint32_t flags;
 };
 
 struct __attribute__((__packed__)) osd_reply_sec_rw_t
@@ -109,6 +111,9 @@ struct __attribute__((__packed__)) osd_op_sec_del_t
     object_id oid;
     // delete version (automatic or specific)
     uint64_t version;
+    // the only possible flag is OSD_OP_RECOVERY_RELATED
+    uint32_t flags;
+    uint32_t pad0;
 };
 
 struct __attribute__((__packed__)) osd_reply_sec_del_t
@@ -121,6 +126,9 @@ struct __attribute__((__packed__)) osd_reply_sec_del_t
 struct __attribute__((__packed__)) osd_op_sec_sync_t
 {
     osd_op_header_t header;
+    // the only possible flag is OSD_OP_RECOVERY_RELATED
+    uint32_t flags;
+    uint32_t pad0;
 };
 
 struct __attribute__((__packed__)) osd_reply_sec_sync_t
@@ -134,6 +142,9 @@ struct __attribute__((__packed__)) osd_op_sec_stab_t
     osd_op_header_t header;
     // obj_ver_id array length in bytes
     uint64_t len;
+    // the only possible flag is OSD_OP_RECOVERY_RELATED
+    uint32_t flags;
+    uint32_t pad0;
 };
 typedef osd_op_sec_stab_t osd_op_sec_rollback_t;
 

src/osd_primary_subops.cpp

@@ -221,6 +221,7 @@ int osd_t::submit_primary_subop_batch(int submit_type, inode_t inode, uint64_t o
                     .offset = wr ? si->write_start : si->read_start,
                     .len = subop_len,
                     .attr_len = wr ? clean_entry_bitmap_size : 0,
+                    .flags = cur_op->peer_fd == SELF_FD && cur_op->req.hdr.opcode != OSD_OP_SCRUB ? OSD_OP_RECOVERY_RELATED : 0,
                 };
 #ifdef OSD_DEBUG
                 printf(
@@ -300,7 +301,8 @@ void osd_t::handle_primary_bs_subop(osd_op_t *subop)
             " retval = "+std::to_string(bs_op->retval)+")"
         );
     }
-    add_bs_subop_stats(subop);
+    bool recovery_related = cur_op->peer_fd == SELF_FD && cur_op->req.hdr.opcode != OSD_OP_SCRUB;
+    add_bs_subop_stats(subop, recovery_related);
     subop->req.hdr.opcode = bs_op_to_osd_op[bs_op->opcode];
     subop->reply.hdr.retval = bs_op->retval;
     if (bs_op->opcode == BS_OP_READ || bs_op->opcode == BS_OP_WRITE || bs_op->opcode == BS_OP_WRITE_STABLE)
@@ -312,30 +314,33 @@ void osd_t::handle_primary_bs_subop(osd_op_t *subop)
     }
     delete bs_op;
     subop->bs_op = NULL;
-    subop->peer_fd = -1;
+    subop->peer_fd = SELF_FD;
-    handle_primary_subop(subop, cur_op);
+    if (recovery_related && recovery_target_sleep_us)
+    {
+        tfd->set_timer_us(recovery_target_sleep_us, false, [=](int timer_id)
+        {
+            handle_primary_subop(subop, cur_op);
+        });
+    }
+    else
+    {
+        handle_primary_subop(subop, cur_op);
+    }
 }
 
-void osd_t::add_bs_subop_stats(osd_op_t *subop)
+void osd_t::add_bs_subop_stats(osd_op_t *subop, bool recovery_related)
 {
     // Include local blockstore ops in statistics
     uint64_t opcode = bs_op_to_osd_op[subop->bs_op->opcode];
     timespec tv_end;
     clock_gettime(CLOCK_REALTIME, &tv_end);
-    msgr.stats.op_stat_count[opcode]++;
+    uint64_t len = (opcode == OSD_OP_SEC_READ || opcode == OSD_OP_SEC_WRITE)
-    if (!msgr.stats.op_stat_count[opcode])
+        ? subop->bs_op->len : 0;
+    msgr.inc_op_stats(msgr.stats, opcode, subop->tv_begin, tv_end, len);
+    if (recovery_related)
     {
-        msgr.stats.op_stat_count[opcode] = 1;
+        // It is OSD_OP_RECOVERY_RELATED
-        msgr.stats.op_stat_sum[opcode] = 0;
+        msgr.inc_op_stats(msgr.recovery_stats, opcode, subop->tv_begin, tv_end, len);
-        msgr.stats.op_stat_bytes[opcode] = 0;
-    }
-    msgr.stats.op_stat_sum[opcode] += (
-        (tv_end.tv_sec - subop->tv_begin.tv_sec)*1000000 +
-        (tv_end.tv_nsec - subop->tv_begin.tv_nsec)/1000
-    );
-    if (opcode == OSD_OP_SEC_READ || opcode == OSD_OP_SEC_WRITE)
-    {
-        msgr.stats.op_stat_bytes[opcode] += subop->bs_op->len;
     }
 }
 
@@ -558,6 +563,7 @@ void osd_t::submit_primary_del_batch(osd_op_t *cur_op, obj_ver_osd_t *chunks_to_
                 },
                 .oid = chunk.oid,
                 .version = chunk.version,
+                .flags = cur_op->peer_fd == SELF_FD && cur_op->req.hdr.opcode != OSD_OP_SCRUB ? OSD_OP_RECOVERY_RELATED : 0,
             } };
             subops[i].callback = [cur_op, this](osd_op_t *subop)
             {
@@ -615,6 +621,7 @@ int osd_t::submit_primary_sync_subops(osd_op_t *cur_op)
                     .id = msgr.next_subop_id++,
                     .opcode = OSD_OP_SEC_SYNC,
                 },
+                .flags = cur_op->peer_fd == SELF_FD && cur_op->req.hdr.opcode != OSD_OP_SCRUB ? OSD_OP_RECOVERY_RELATED : 0,
             } };
             subops[i].callback = [cur_op, this](osd_op_t *subop)
             {
@@ -674,6 +681,7 @@ void osd_t::submit_primary_stab_subops(osd_op_t *cur_op)
                     .opcode = OSD_OP_SEC_STABILIZE,
                 },
                 .len = (uint64_t)(stab_osd.len * sizeof(obj_ver_id)),
+                .flags = cur_op->peer_fd == SELF_FD && cur_op->req.hdr.opcode != OSD_OP_SCRUB ? OSD_OP_RECOVERY_RELATED : 0,
             } };
             subops[i].iov.push_back(op_data->unstable_writes + stab_osd.start, stab_osd.len * sizeof(obj_ver_id));
             subops[i].callback = [cur_op, this](osd_op_t *subop)

src/osd_primary_write.cpp

@@ -296,7 +296,6 @@ resume_7:
             if (!recovery_stat[recovery_type].count) // wrapped
             {
                 memset(&recovery_print_prev[recovery_type], 0, sizeof(recovery_print_prev[recovery_type]));
-                memset(&rtune_prev_recovery[recovery_type], 0, sizeof(rtune_prev_recovery[recovery_type]));
                 memset(&recovery_stat[recovery_type], 0, sizeof(recovery_stat[recovery_type]));
                 recovery_stat[recovery_type].count++;
             }

src/osd_secondary.cpp

@@ -42,7 +42,21 @@ void osd_t::secondary_op_callback(osd_op_t *op)
     int retval = op->bs_op->retval;
     delete op->bs_op;
     op->bs_op = NULL;
-    finish_op(op, retval);
+    if (op->is_recovery_related() && recovery_target_sleep_us)
+    {
+        if (!op->tv_end.tv_sec)
+        {
+            clock_gettime(CLOCK_REALTIME, &op->tv_end);
+        }
+        tfd->set_timer_us(recovery_target_sleep_us, false, [this, op, retval](int timer_id)
+        {
+            finish_op(op, retval);
+        });
+    }
+    else
+    {
+        finish_op(op, retval);
+    }
 }
 
 void osd_t::exec_secondary(osd_op_t *cur_op)