Viewing a response to: @steemitblog/steemit-blockchain-team-update-appbase-rocksdb-bandwidth-hf20-smts-and-more
as you mentioned, Rocksdb is heavily optimized for machines with SSD/flash storage because it has a mmtable + sstable approach where writes first go to the mmtable and batched writes go to the sstable. this process can be write through as well. doesn't this approach necessitate presence of SSD drives all the time just because of the expectation of minimal persistence latencies. am i correct in assuming that we are okay with this approach because only the witnesses need to run rocksdb in combination with reasonable RAM and higher SSD capacities? thanks
author | adarshh |
---|---|
permlink | re-steemitblog-steemit-blockchain-team-update-appbase-rocksdb-bandwidth-hf20-smts-and-more-20180524t163143423z |
category | steem |
json_metadata | {"tags":["steem"],"app":"steemit/0.1"} |
created | 2018-05-24 16:31:45 |
last_update | 2018-05-24 16:31:45 |
depth | 1 |
children | 0 |
last_payout | 2018-05-31 16:31:45 |
cashout_time | 1969-12-31 23:59:59 |
total_payout_value | 0.000 HBD |
curator_payout_value | 0.000 HBD |
pending_payout_value | 0.000 HBD |
promoted | 0.000 HBD |
body_length | 564 |
author_reputation | 1,909,676,117,380 |
root_title | "Steemit Blockchain Team Update: AppBase, RocksDB, Bandwidth, HF20, SMTs, and more!" |
beneficiaries | [] |
max_accepted_payout | 1,000,000.000 HBD |
percent_hbd | 10,000 |
post_id | 57,485,847 |
net_rshares | 0 |