Index batch ingestion takes to much time to finish

Hi, List

I’m trying to ingest data (765M) by index batch ingestion on Druid 0.8.3. The ingestion spec has 43 dimensions and 4012 metrics. I’m running this on overlord on dedicated machine (with no middleManager):

Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 12
On-line CPU(s) list: 0-11
Thread(s) per core: 1
Core(s) per socket: 1
Socket(s): 12
NUMA node(s): 1
Vendor ID: GenuineIntel
CPU family: 6
Model: 45
Model name: Intel® Xeon® CPU E5-2440 0 @ 2.40GHz
Stepping: 7
CPU MHz: 2399.998
BogoMIPS: 4799.99
Virtualization: VT-x
Hypervisor vendor: KVM
Virtualization type: full
L1d cache: 32K
L1i cache: 32K
L2 cache: 256K
L3 cache: 15360K
NUMA node0 CPU(s): 0-11

Memory: 32GB

My overlord in running on local mode and I also set on properties:
druid.indexer.runner.javaOpts=-server -Xmx28g -XX:-UseGCOverheadLimit

This task takes to much time to finish (over than 1h). Is there something that I can do to improve the task performance? The time is related with the number of metrics (4012)?

Are you able to run batch ingestion through EMR or a remote Hadoop cluster?