Overview of DOcDB
AUTOMATIC, CONTINUOUS, incremental backups and PITR
T
docdb stores JSON documents (semi structured data)
T
key vaue pairs can be nested
T
why docdb?
run ad hoc queries for operational and analytics workloads
for known access patterns - use dynamoDB instead
T
DocDB architecture
___ copies of your data across ___ AZs
6, 3
DocDB Architecture
_____ optimistic algorithm (quorum model)
lock-free
docdb architecture
___ copies out of 6 needed for writes (__/6 write quorum - data considered durable whe at least __/6 copies acknowledge the write)
4
docdb architecture
___ copies out of 6 needed for reads (__/6 read quorum
3
docdb architecture
self healing with ____ replication, storage is striped across 100s of volumes
peer to peer
docdb architecture
___ docdb isntance takes writes(master)
1
___ nodes on replicas do not need to write/replicate (=improved read performance)
compute
docdb architecture
log-structured distributed storage layer - passes ___ log records from compute to stroage layer (=faster)
incremental
docdb architecture
master + up to ___ read replicas server reads
15
docdb architecture
data is continuously backed up to __ in real time, using storage nodes (compute nore perfomance is unaffected)
s3
docdb cluster
T
docdb replication
15
docdb replication
____ repliaction
ASYNC
docdb replication
replicas share the same underlying storage layer
T
docdb replication
typically take 10s of milliseconds (replication lag)
T
docdb replication
minimal performance impact on the primary due to replication process
T
docdb HA failovers
-failovers occur automatically
T
docdb HA failovers
-a replica is automatically promoted to be the new primary during DR
T
docdb HA failovers
docdb flops the CNAME of the DB instance to point ot the replica and promotes it
T
docdb HA failovers
failover to a replica typically takes 30 seconds ( minimal downtime)
t