Reduce ds_id name length (microsoft#5176)

Fixing issue microsoft#5087 . Limited the naming of the ds_id in ZeRO 3 to the
first and last parameters of the group instead of every parameter in the
group.

deepspeed/runtime/zero/stage3.py

@@ -864,7 +864,9 @@ def _create_fp32_partitions(self):
                             self.device).clone().float().detach())
 
             self.fp32_partitioned_groups_flat[i].requires_grad = True  # keep this in case internal optimizer uses it
-            self.fp32_partitioned_groups_flat[i].ds_id = '_'.join(map(str, self.fp16_partitioned_groups_flat_id[i]))
+            ds_id_begin = str(self.fp16_partitioned_groups_flat_id[i][0])
+            ds_id_end = str(self.fp16_partitioned_groups_flat_id[i][-1])
+            self.fp32_partitioned_groups_flat[i].ds_id = ds_id_begin + '_' + ds_id_end
 
         if len(swappable_fp32_tensors) > 0:
             self.optimizer_swapper.initialize_parameters(parameters=swappable_fp32_tensors,