Scheduling snapshots
There are disclosed computer-implemented methods, apparatus, and computer program products for scheduling snapshots. In one embodiment, the method comprises the following steps. The method comprises receiving performance data relating to a data storage system. The method also comprises determining, based on the performance data, a time for performing a snapshot of data stored on the data storage system. The method further comprises scheduling the snapshot according to the time.
1. A computer-implemented method, comprising:
receiving statistics data representing historic performance statistics over a predetermined period of time by a storage system, the historic performance statistics including resource consumption of a plurality of resources including at least one of a processor, memory, input-output (IO) transactions, and network bandwidth;
performing an analysis, by an analysis module executed by a processor, on the historic performance statistics to determine an optimal time window within the predetermined time period for taking a snapshot of data stored on the storage system based on the analysis, wherein performing an analysis on the historic performance statistics comprises assigning a weight factor for each resource for each point in time over the predetermined time period during which the historic performance statistics were collected, ranking each resource based on its resource consumption in view of its corresponding weight factor for each point in time over the predetermined time period, calculating a standard mean value and a standard deviation value based on resource rankings of sample points in time for each of a plurality of time window candidates, and selecting a time window from the time window candidates for taking snapshot, such that during the selected time window the standard mean value is within a first predetermined range while the standard deviation is within a second predetermined range; and
scheduling, by a scheduler executed by the processor, the snapshot according to the optimal time window.
2. An apparatus, comprising:
a processor; and
a memory coupled to the processor for storing instructions, which when executed from the memory, cause the processor to:
receive statistics data representing historic performance statistics over a predetermined period of time by a storage system, the historic performance statistics including resource consumption of a plurality of resources including at least one of a processor, memory, input-output (IO) transactions, and network bandwidth;
perform an analysis, by an analysis module executed by processor, on the historic performance statistics to determine an optimal time window within the predetermined time period for taking a snapshot of data stored on the storage system based on the analysis, wherein performing an analysis on the historic performance statistics comprises assigning a weight factor for each resource for each point in time over the predetermined time period during which the historic performance statistics were collected, ranking each resource based on its resource consumption in view of its corresponding weight factor for each point in time over the predetermined time period, calculating a standard mean value and a standard deviation value based on resource rankings of sample points in time for each of a plurality of time window candidates, and selecting a time window from the time window candidates for taking snapshot, such that during the selected time window the standard mean value is within a first predetermined range while the standard deviation is within a second predetermined range; and
schedule, by a scheduler executed by processor, the snapshot according to the optimal time window.
3. A computer program product having a non-transitory computer-readable medium storing instructions, the instructions, when carried out by a processor, causing the processor to perform a method of:
receiving statistics data representing historic performance statistics over a predetermined period of time by a storage system, the historic performance statistics including resource consumption of a plurality of resources including at least one of a processor, memory, input-output (IO) transactions, and network bandwidth;
performing an analysis, by an analysis module executed by processor, on the historic performance statistics to determine an optimal time window within the predetermined time period for taking a snapshot of data stored on the storage system based on the analysis, wherein performing an analysis on the historic performance statistics comprises assigning a weight factor for each resource for each point in time over the predetermined time period during which the historic performance statistics were collected, ranking each resource based on its resource consumption in view of its corresponding weight factor for each point in time over the predetermined time period, calculating a standard mean value and a standard deviation value based on resource rankings of sample points in time for each of a plurality of time window candidates, and selecting a time window from the time window candidates for taking snapshot, such that during the selected time window the standard mean value is within a first predetermined range while the standard deviation is within a second predetermined range; and
scheduling, by a scheduler executed by processor, the snapshot according to the optimal time window.
4. A computer-implemented method, comprising:
receiving statistics data representing historic performance statistics over a predetermined period of time by a storage system, the historic performance statistics including resource consumption of a plurality of resources including at least one of a processor, memory, input-output (IO) transactions, and network bandwidth;
performing an analysis, by an analysis module executed by a processor, on the historic performance statistics to determine an optimal time window within the predetermined time period for taking a snapshot of data stored on the storage system based on the analysis, wherein performing an analysis on the historic performance statistics comprises assigning a weight factor for each resource for each point in time over the predetermined time period during which the historic performance statistics were collected, and ranking each resource based on its resource consumption in view of its corresponding weight factor for each point in time over the predetermined time period, wherein ranking r i (t) for resource i at time t is determined by r i (t)=w i (t)*(M i -v i (t))/(M i −N i ), wherein w i (t) is a weight factor for resource i at time t, wherein v i (t) is a consumption level of resource i at time t, wherein M i is a maximum value at time t and N i is a minimum value at time t; and
scheduling, by a scheduler executed by the processor, the snapshot according to the optimal time window.
5. An apparatus, comprising:
a processor; and
a memory coupled to the processor for storing instructions, which when executed from the memory, cause the processor to:
receive statistics data representing historic performance statistics over a predetermined period of time by a storage system, the historic performance statistics including resource consumption of a plurality of resources including at least one of a processor, memory, input-output (IO) transactions, and network bandwidth;
perform an analysis, by an analysis module executed by processor, on the historic performance statistics to determine an optimal time window within the predetermined time period for taking a snapshot of data stored on the storage system based on the analysis, wherein performing an analysis on the historic performance statistics comprises assigning a weight factor for each resource for each point in time over the predetermined time period during which the historic performance statistics were collected, and ranking each resource based on its resource consumption in view of its corresponding weight factor for each point in time over the predetermined time period, wherein ranking r i (t) for resource i at time t is determined by r i (t)=w i (t)*(M i −v i (t))/(M i −N i ), wherein w i (t) is a weight factor for resource i at time t, wherein v i (t) is a consumption level of resource i at time t, wherein M i is a maximum value at time t and N i is a minimum value at time t; and
schedule, by a scheduler executed by processor, the snapshot according to the optimal time window.
6. A computer program product having a non-transitory computer-readable medium storing instructions, the instructions, when carried out by a processor, causing the processor to perform a method of:
receiving statistics data representing historic performance statistics over a predetermined period of time by a storage system, the historic performance statistics including resource consumption of a plurality of resources including at least one of a processor, memory, input-output (IO) transactions, and network bandwidth;
performing an analysis, by an analysis module executed by processor, on the historic performance statistics to determine an optimal time window within the predetermined time period for taking a snapshot of data stored on the storage system based on the analysis, wherein performing an analysis on the historic performance statistics comprises assigning a weight factor for each resource for each point in time over the predetermined time period during which the historic performance statistics were collected, and ranking each resource based on its resource consumption in view of its corresponding weight factor for each point in time over the predetermined time period, wherein ranking r i (t) for resource i at time t is determined by r i (t)=w i (t)*(M i −v i (t))/(M i −N i ), wherein w i (t) is a weight factor for resource i at time t, wherein v i (t) is a consumption level of resource i at time t, wherein M i is a maximum value at time t and N i is a minimum value at time t; and
scheduling, by a scheduler executed by processor, the snapshot according to the optimal time window.
7. A computer-implemented method, comprising:
receiving statistics data representing historic performance statistics over a predetermined period of time by a storage system, the historic performance statistics including resource consumption of a plurality of resources including at least one of a processor, memory, input-output (IO) transactions, and network bandwidth;
performing an analysis, by an analysis module executed by a processor, on the historic performance statistics to determine an optimal time window within the predetermined time period for taking a snapshot of data stored on the storage system based on the analysis, wherein performing an analysis on the historic performance statistics comprises assigning a weight factor for each resource for each point in time over the predetermined time period during which the historic performance statistics were collected, and ranking each resource based on its resource consumption in view of its corresponding weight factor for each point in time over the predetermined time period, wherein ranking r i (t) for resource i at time t is determined by r i (t)=w i (t)*exp(−v i (t) 2 /2σ 2 ), wherein w i (t) is a weight factor for resource i at time t, wherein v i (t) is a consumption level of resource i at time t, and wherein σ is calculated as a standard deviation for a discrete random variable; and
scheduling, by a scheduler executed by the processor, the snapshot according to the optimal time window.
8. An apparatus, comprising:
a processor; and
a memory coupled to the processor for storing instructions, which when executed from the memory, cause the processor to:
receive statistics data representing historic performance statistics over a predetermined period of time by a storage system, the historic performance statistics including resource consumption of a plurality of resources including at least one of a processor, memory, input-output (IO) transactions, and network bandwidth;
perform an analysis, by an analysis module executed by processor, on the historic performance statistics to determine an optimal time window within the predetermined time period for taking a snapshot of data stored on the storage system based on the analysis, wherein performing an analysis on the historic performance statistics comprises assigning a weight factor for each resource for each point in time over the predetermined time period during which the historic performance statistics were collected, and ranking each resource based on its resource consumption in view of its corresponding weight factor for each point in time over the predetermined time period, wherein ranking r i (t) for resource i at time t is determined by r i (t)=w i (t)*exp(−v i (t) 2 /2σ 2 ), wherein w i (t) is a weight factor for resource i at time t, wherein v i (t) is a consumption level of resource i at time t, and wherein σ is calculated as a standard deviation for a discrete random variable; and
schedule, by a scheduler executed by processor, the snapshot according to the optimal time window.
9. A computer program product having a non-transitory computer-readable medium storing instructions, the instructions, when carried out by a processor, causing the processor to perform a method of:
receiving statistics data representing historic performance statistics over a predetermined period of time by a storage system, the historic performance statistics including resource consumption of a plurality of resources including at least one of a processor, memory, input-output (IO) transactions, and network bandwidth;
performing an analysis, by an analysis module executed by processor, on the historic performance statistics to determine an optimal time window within the predetermined time period for taking a snapshot of data stored on the storage system based on the analysis, wherein performing an analysis on the historic performance statistics comprises assigning a weight factor for each resource for each point in time over the predetermined time period during which the historic performance statistics were collected, and ranking each resource based on its resource consumption in view of its corresponding weight factor for each point in time over the predetermined time period, wherein ranking r i (t) for resource i at time t is determined by r i (t)=w i (t)*exp(−v i (t) 2 /2σ 2 ), wherein w i (t) is a weight factor for resource i at time t, wherein v i (t) is a consumption level of resource i at time t, and wherein σ is calculated as a standard deviation for a discrete random variable; and
scheduling, by a scheduler executed by processor, the snapshot according to the optimal time window.