Encrypted Replication

Print this Page	2.6 Business Continuity 04 Encrypted Replication	Close this Page

Strategic Board	Design Division	Transition Division	Operations Division	Improvement Division			ITIL-V3 Bookcase	Table of Contents	Close this Page

Service Catalogue Manager	Service Level Manager	Risk Manager	Capacity Manager	Availability Manager	Business Continuity Manager	Infoformation Security Manager	Compliance Manager		Design Division

ITIL 2.6.2 Business Continuity Plan

Strategy for a better ERA:

1. Bespoke Application Services have evolved to incorporate three privacy and security factors as:-

(1) Encryption so data cannot be stolen.

(2) Replication so data cannot be lost.

(3) Authentication so data cannot be hacked.

2. Encryption of 100% of all business data has been a massive improvement that prevents data from being stolen.

3. Replication of all business data has been a major improvement that prevents application downtime.

4. Authentication has evolved into fingerprinting computer devices so approved people can be identified and criminals cannot impersonate an approved person.

Encryption:

1. Business data is never stored in a readable form and it is plausible to state that business data is not stored because nobody could ever prove that business data was stored. 100% of all business data is encrypted using many layers of different encryption methods so if one layer is cracked, other layers continue to encrypt the information.

2. Encrypted data uses pseudonymisation as a principal method where field values are replaced with tokens that cannot be reverse engineered. Encrypted data is meaningless and stored in photographs and images that are hidden in a library of more than a million similar photographs and images.

3. Encrypted has evolved with artificial intelligence with more and more layers of encryption methods that may not be cracked by applying more and more computing power. Fake data and real business data are entwined with many plausible results of decryption - it may be impossible for a criminal to know what is fake data.

4. SQL databases have been replaced with photographs and images that contain encrypted data so SQL injection attacks have been eliminated.

Replication:

1. Once upon a time, backups were taken each evening with the hope that in the event of a disaster, a new database could be recreated from the backup and daily transaction log would reapply all work done since the last backup. In practice, the majority of backup-recovery plans failed because they were untestable and transactions cannot be applied to simulate a full days work for many hundreds of people working in parallel. The majority of criminal thefts of data were from backups that were not encrypted and had to be physically moved to other locations - backups are an expensive liability. It is suspected that many companies had copies of their backup data stolen and they never noticed.

2. Two evolutionary improvements had to be made:-

(1) Replicate encrypted data in a large number of distributed data centers in real-time.

(2) Use an artificial intelligent assistant that never stops to replace application programs that will fail from time to time.

3. The benefit was that:-

(1) downtime was eliminated.

(2) application program downloads and patching was eliminated.

(3) data cannot be stolen.

(4) data cannot be lost.

4. In the event that a data center is not available, business continues to be provided by another data center using replicated data.

Glossary:

PDC means "Primary Data Center" that is providing the Bespoke Application Service (BAS).

SDC means "Secondary Data Center" that may be used to provide the Bespoke Application Service (BAS).

AMQ means "Asynchronous Message Queue" of image names in a PDC for each SDC.

RFM means "Request for Message" transaction from SDC to PDC.

CMT means "Confirm Message Transfer" transaction from SDC to PDC.

AM means "Available Message" reply with data from PDC to SDC.

EM means "Empty Message" reply from PDC to SDC.

Replication Framework:

1. Each PDC has a large number of SDC acting as fallback.

2. The PDC has a unique AMQ for each SDC. An AMQ is a first-in-first-out list of image names.

3. Each SDC is responsible to replicate data from its PDC using RFD and CDT transactions. The PDC is not responsible for replicating data to a SDC. The PDC only responds to transactions from each SDC.

4. When a SDC is not available for some time, the PDC will build up a AMQ of image names that need to be replicated to the SDC when it becomes available.

5. The transaction rate of each SDC will vary depending on the day, hour and size of message queue. If a message queue has built up with a communication delay, then transactions will be continuous until the message queue is empty.

6. The unit of replication is an image file that may contain one or many records. Generally, a large number of small image files are designed.

How does it work:

0. The SDC sends a RFM transaction to the PDC.

1. When the PDC gets a RFM, it checks its AMQ for the SDC and will either:-

(1) Reply with a AM message and applicable image.

(2) Reply with a EM message.

2. When the SDC gets the AM mesage and image:-

(1) The image is stored in the image library (insert or replace).

(2) The SDC sends a CDT transaction with image name to the PDC.

3. When the PDC gets a CDT transaction with image name:-

(1) The message is removed from the AMQ for that SDC.

(2) The SDC sends a RFM transaction to the PDC.

4. When the SDC gets the EM mesage it will:-

(1) Wait for a number of seconds depending on the day and hour.

(2) The SDC sends a RFM transaction to the PDC.

Document Control:

1. Document Title: Encrypted Replication.

2. Reference: 162604.

3. Keywords: ITIL, Encrypted Replication.

4. Description: Encrypted Replication Authentication (ERA).

5. Privacy: Public education service as a benefit to humanity.

6. Issued: 11 Nov 2017.

7. Edition: 1.2.