CAP Model

CAP model

Introduction

Large distributed systems run into a problem which smaller systems don’t usually have to worry about. “Brewers CAP Theorem” defines this problem in a very simple way. It states, that though its desirable to have Consistency, High-Availability and Partition-tolerance in every system, unfortunately no system can achieve all three at the same time.

Following the CAP theorem OrientDB supports CA while OrientKV supports AP.

Consistency

All nodes see the same data at the same time.

"A fully Consistent system is one where the system can guarantee that once you store a state (lets say “x=y”) in the system, it will report the same state in every subsequent operation until the imagestate is explicitly changed by something outside the system."

Availability

Node failures do not prevent survivors from continuing to operate.

Partition-tolerance

The system continues to operate despite arbitrary message loss.

CAP model matrix

Usage	**(C)onsistency**	**(A)vailability**	**(P)artition-tolerance**
Local database	Yes	No	No
Remote database	Yes	No	No
Remote clustered database	Yes	Yes	No
Key/Value server	No	Yes	Yes

Disk-write strategies

Orient, by default, is configured for synchronous writes but depending by the OS file system management changes can be permanents only after the OS flushes the buffer into the disk. If you're using a transaction you're protect in case any crash happens when data are not written to the disk yet.

For this reason Orient permits to change this policy.

Synchronize all changes to the disk when any single-record is changed

By default is false.

    database.getStorage().getConfiguration().txSegment.setSynchRecord(true);

Synchronize all changes to the disk when a transaction is committed

By default is true.

    database.getStorage().getConfiguration().txSegment.setSynchTx(true);