increase thread count and correct mode printed to outfile

qdp_project/src/benchmark/MAX_benchmark.cpp

@@ -127,7 +127,7 @@ int main(int argc, char** argv) {
     Linear_Int_Range<uint32_t, 0, 30, 1> run("run");
     Linear_Int_Range<size_t, chunk_min, chunk_max, chunk_incr> chunk_size("chunk_size");
 
-    print_to_file(out_file, generateHead(run, chunk_size, "HBM-Prefetch"), "thread_group", "time",
+    print_to_file(out_file, generateHead(run, chunk_size, "mode"), "thread_group", "time",
     #ifdef THREAD_TIMINGS
         "scan_a", "scan_b", "aggr_j",
     #endif
@@ -175,9 +175,9 @@ int main(int argc, char** argv) {
         std::promise<void> p;
         std::shared_future<void> ready_future(p.get_future());
 
-        const uint8_t tc_filter = 4;
-        const uint8_t tc_copy   = 1;
-        const uint8_t tc_agg    = 2;
+        const uint8_t tc_filter = 6;
+        const uint8_t tc_copy   = 2;
+        const uint8_t tc_agg    = 4;
 
         Query_Wrapper<base_t, simple_query, cache_a, wait_b> qw (
                 &ready_future, workload_b, chunk_size.current,
@@ -195,13 +195,6 @@ int main(int argc, char** argv) {
         std::vector<std::thread>   copy_pool;
         std::vector<std::thread>    agg_pool;
 
-        // int thread_id = 0;
-        // std::vector<std::pair<int, int>> pinning_ranges {std::make_pair(28, 42), std::make_pair(84, 98)}; // node 2 heacboehm II
-        //std::vector<std::pair<int, int>> pinning_ranges {std::make_pair(32, 48), std::make_pair(96, 112)}; // node 2 heacboehm
-        // std::vector<std::pair<int, int>> pinning_ranges {std::make_pair(24, 36), std::make_pair(120, 132)}; // node 2 sapphire rapids
-        //std::vector<std::pair<int, int>> pinning_ranges {std::make_pair(24, 48)}; // node 2+3 sapphire rapids
-        //std::vector<std::pair<int, int>> pinning_ranges {std::make_pair(0, 48)}; // node 0-3 sapphire rapids
-
         for(uint32_t gid = 0; gid < THREAD_GROUP_MULTIPLIER; ++gid) {
             for(uint32_t tid = 0; tid < tc_filter; ++tid) {
                 filter_pool.emplace_back(filter_lambda, gid, THREAD_GROUP_MULTIPLIER, tid);
@@ -232,7 +225,7 @@ int main(int argc, char** argv) {
         double seconds   = (double)(nanos) / nanos_per_second;
 
         if (i >= 5) {
-            print_to_file(out_file, run, chunk_size, "HBM", THREAD_GROUP_MULTIPLIER, seconds,
+            print_to_file(out_file, run, chunk_size, "DSA-HBM-Prefetch", THREAD_GROUP_MULTIPLIER, seconds,
                 #ifdef THREAD_TIMINGS
                           qw.trt->summarize_time(0),  qw.trt->summarize_time(1),  qw.trt->summarize_time(2),
                 #endif