9 billion entry db: client crashes

[robertkowalski]

with noise network small databases work like a charm. with a large db (107GB, 9 billion entries) the protocol crashes.

steps to reproduce:

the write test in the readme creates a database with 9 billion entries

the other test, the p2p test in the readme works for small databases, but with a large db, they all get a timeout, so basically no client can connect:

further infos: see readme in https://github.com/bitfinexcom/hypercore-benchmark to setup test db and noise benchmark.

stacktrace:

replication ended Error: Remote timed out
    at Protocol._ontimeout (/mnt/storage-hypercore/hypercore-test/hypertest/node_modules/hypercore-protocol/index.js:453:16)
    at Object.onceWrapper (events.js:273:13)
    at Protocol.emit (events.js:182:13)
    at Protocol._kick (/mnt/storage-hypercore/hypercore-test/hypertest/node_modules/hypercore-protocol/index.js:196:10)
    at Timeout.kick (/mnt/storage-hypercore/hypercore-test/hypertest/node_modules/hypercore-protocol/index.js:96:10)
    at ontimeout (timers.js:436:11)
    at tryOnTimeout (timers.js:300:5)
    at unrefdHandle (timers.js:520:7)
    at Timer.processTimers (timers.js:222:12)