Application service-level configuration of dataloss failover

A computer-implemented method performed by one or more processors executing computer-executable instructions, wherein the computer-implemented method manages at a data-tier level failover for application services that are each serviced at the data-tier level by a respective primary data store and a respective secondary data store used to back up the respective primary data store during a failover, the computer-implemented method comprising:

• each application service communicating through an application program interface to a management component at the data-tier level, wherein said communication comprises defining a failover policy for each application service and wherein the failover policy is defined by:
• a dataloss failover parameter that defines at what point in time, in case of at least a partial failure of a respective primary data store of a respective application service, the respective application service should perform a dataloss failover to a respective secondary data store of the respective application service;
• and
• a dataloss-avoidant parameter that defines at what point in time, in the case of the at least partial failure of the respective primary data store of the respective application service, the respective application service should attempt a lossless failover to the respective secondary data store of the respective application service;
• the management component storing at the data-tier level the dataloss failover parameter and the dataloss-avoidant parameter for each application service;
• following a failure of the respective primary data store for a given application service, using the failover policy stored for the given application service to detect the following conditions:
• first, whether, at the point in time defined by the dataloss-avoidant parameter, lossless recovery of the respective primary data store for the given application service has not been accomplished;
• and
• second, whether, at the point in time defined by the dataloss failover parameter, the respective primary data store for the given application service is at least still partially unavailable to the given application service;
• and
• if the first and second conditions are both detected, thereafter performing a dataloss failover from the respective primary data store to the respective secondary data store of the given application service.

A computer-implemented method performed by one or more processors executing computer-executable instructions, wherein the computer-implemented method manages at a data-tier level failover for application services that are each serviced at the data-tier level by a respective primary data store and a respective secondary data store used to back up the respective primary data store during a failover, the computer-implemented method comprising:

• receiving at a management component at the data-tier level a failover policy defined by each application service, wherein the failover policy is communicated to the management component by each application service through an application program interface for the data-tier level, wherein said communication comprises defining a failover policy for each application service and wherein the failover policy is defined by:
• a dataloss failover parameter that defines at what point in time, in case of at least a partial failure of a respective primary data store of a respective application service, the respective application service should perform a dataloss failover to a respective secondary data store of the respective application service;
• and
• a dataloss-avoidant parameter that defines at what point in time, in the case of the at least partial failure of the respective primary data store of the respective application service, the respective application service should attempt a lossless failover to the respective secondary data store of the respective application service;
• the management component storing at the data-tier level the dataloss failover parameter and the dataloss-avoidant parameter for each application service;
• following a failure of the respective primary data store for a given application service, the management component using the failover policy stored for the given application service to detect the following conditions:
• first, whether, at the point in time defined by the dataloss-avoidant parameter, lossless recovery of the respective primary data store for the given application service has not been accomplished;
• and
• second, whether, at the point in time defined by the dataloss failover parameter, the respective primary data store for the given application service is at least still partially unavailable to the given application service;
• and
• if the first and second conditions are both detected, thereafter the management component performing a dataloss failover from the respective primary data store to the respective secondary data store of the given application service so that in accordance with the failover policy defined by the given application service, any further latency resulting from the dataloss failover from the respective primary data store to the respective secondary data store will be reduced irrespective of any likely data loss that may occur during the dataloss failover.