4 3 2 1 Application Data Protection Availability

This is part two of a five-part mini-series looking at Application Data Value Characteristics everything is not the same across IT data centers, data infrastructures including cloud environments. In this post, we continue looking at application performance, availability, capacity, economic (PACE) attributes that have an impact on data value as well as availability.

Availability (Accessibility, Durability, Consistency)

Just as there are many different aspects and focus areas for performance, there are also several facets to availability. Note that applications performance requires availability and availability relies on some level of performance.

Availability is a broad and encompassing area that includes data protection to protect, preserve, and serve (backup/restore, archive, BC, BR, DR, HA) data and applications. There are logical and physical aspects of availability including data protection as well as security including key management (manage your keys or authentication and certificates) and permissions, among other things.

Availability = accessibility (can you get to your application and data) + durability (is the data intact and consistent). This includes basic Reliability, Availability, Serviceability (RAS), as well as high availability, accessibility, and durability. “Durable” has multiple meanings, so context is important. Durable means how data infrastructure resources hold up to, survive, and tolerate wear and tear from use (i.e., endurance), for example, Flash SSD or mechanical devices such as Hard Disk Drives (HDDs). Another context for durable refers to data, meaning how many copies in various places.

From a data protection standpoint, the fundamental rule or guideline is 4 3 2 1, which means having at least four copies consisting of at least three versions (different points in time), at least two of which are on different systems or storage devices and at least one of those is off-site (on-line, off-line, cloud, or other). There are many variations of the 4 3 2 1 rule shown in the following figure along with approaches on how to manage technology to use. We will go into deeper this subject in later chapters. For now, remember the following.

4  At least four copies of data (or more), Enables durability in case a copy goes bad, deleted, corrupted, failed device, or site.

3    The number (or more) versions of the data to retain, Enables various recovery points in time to restore, resume, restart from.

2    Data located on two or more systems (devices or media/mediums), Enables protection against device, system, server, file system, or other fault/failure.

1   With at least one of those copies being off-premise and not live (isolated from active primary copy), Enables resiliency across sites, as well as space, time, distance gap for protection.

What this all means and wrap-up

Keep in mind that with Application Data Value Characteristics Everything Is Not The Same across various organizations, data centers, data infrastructures spanning legacy, cloud and other software defined data center (SDDC) environments. All applications have some element of performance, availability, capacity, economic (PACE) needs as well as resource demands. There is often a focus around data storage about storage efficiency and utilization which is where data footprint reduction (DFR) techniques, tools, trends and as well as technologies address capacity requirements. However with data storage there is also an expanding focus around storage effectiveness also known as productivity tied to performance, along with availability including 4 3 2 1 data protection.

Read more here.