Multi-tenant memory service for memory pool architectures

• US 10,409,509 B2
• Filed: 11/03/2017
• Issued: 09/10/2019
• Est. Priority Date: 04/07/2016
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bivens, John Alan, Das, Koushik K., Li, Min, Mahindru, Ruchi, Ramasamy, Harigovind V., Ruan, Yaoping, Salapura, Valentina, Schenfeld, Eugen
• Primary Examiner(s): Thai, Tuan V
• Application Number:
  US15/803,004
• Publication Number:
  US 20180074741A1
• Time in Patent Office: 676 Days
• CPC Class Codes: G06F 11/1446 Point-in-time backing up or restoration of persistent data
• CPC Class Codes: G06F 11/1453 using de-duplication of the data
• CPC Class Codes: G06F 11/3006 where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems multiprogramming arrangements G06F9/46; allocation of resources G06F9/50
• CPC Class Codes: G06F 11/3433 for load management allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029
• CPC Class Codes: G06F 12/023 Free address space management
• CPC Class Codes: G06F 12/0284 Multiple user address space allocation, e.g. using different base addresses interprocessor communication G06F15/163
• CPC Class Codes: G06F 12/0806 Multiuser, multiprocessor or multiprocessing cache systems
• CPC Class Codes: G06F 12/0808 with cache invalidating means G06F12/0815 takes precedence
• CPC Class Codes: G06F 2212/1032 Reliability improvement, data loss prevention, degraded operation etc
• CPC Class Codes: G06F 2212/152 Virtualized environment, e.g. logically partitioned system
• CPC Class Codes: G06F 2212/154 Networked environment
• CPC Class Codes: G06F 2212/502 using adaptive policy
• CPC Class Codes: G06F 2212/62 Details of cache specific to multiprocessor cache arrangements
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0631 by allocating resources to storage systems
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/0641 De-duplication techniques
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0653 Monitoring storage devices or systems
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

A memory management service occupies a configurable portion of an overall memory system in a disaggregate compute environment. The service provides optimized data organization capabilities over the pool of real memory accessible to the system. The service enables various types of data stores to be implemented in hardware, including at a data structure level. Storage capacity conservation is enabled through the creation and management of high-performance, re-usable data structure implementations across the memory pool, and then using analytics (e.g., multi-tenant similarity and duplicate detection) to determine when data organizations should be used. The service also may re-align memory to different data structures that may be more efficient given data usage and distribution patterns. The service also advantageously manages automated backups efficiently.

Claims

1. A method of managing memory in a multi-tenant compute environment, comprising

providing a memory resource pool;

for a given workload, allocating memory resources from the memory resource pool to a set of processors;

organizing data for the given workload in the allocated memory resources according to a data structure, wherein the data structure is a construct implemented to organize the data and is distinct from the data; and

determining whether re-organization of the data in the data structure for the given workload would increase performance of the data structure or conserve memory used by the data structure; and

upon a determination that re-organization would increase performance or conserve memory, taking a given action to manage the data structure, wherein the given action detects data similarity or duplicates and, in response thereto, eliminates given data from the data structure.

2. The method as described in claim 1 wherein data similarity or duplicates are detected by:

identifying data blocks that are common to multiple tenants or multiple processes;

upon identifying common data blocks, determining whether the common data blocks are stable according to a configurable threshold;

upon a determination that the common data blocks are stable, eliminating duplicate common data blocks;

associating an instance of the common data blocks as a shared common data block; and

providing the multiple tenants or multiple processes a pointer to the shared common data block.

3. The method as described in claim 2, further including:

determining that a value for a shared common data block has changed for a first tenant;

determining whether it is computationally less expensive to store a delta value for the shared common data block for the first tenant; and

upon a determination that it is computationally less expensive, storing the delta value for the shared common data block for the first tenant.

4. The method as described in claim 3, further comprising:

determining whether it is computationally less expensive to store a new value for the shared common data block for the first tenant; and

upon a determination that it is computationally less expensive to store a new value, storing the new value for the shared common data block for the first the tenant.

5. The method as described in claim 1, wherein the given action to manage the data comprises:

monitoring data distribution or usage patterns of the data; and

based on the monitoring, applying an optimization to adjust the data structure.

6. The method as described in claim 1, further including:

profiling a data set to produce a data set schema;

periodically backing up the data set without using a processor; and

flushing cached data to main memory prior to a backup operation.

7. The method as described in claim 1 wherein the data structure organizes the data based on an access requirement of the given workload.

8. An apparatus of managing memory in a multi-tenant compute environment, comprising:

one or more hardware processors;

computer memory holding computer program instructions executed by the hardware processors and operative to;

organize a memory resource pool;

for a given workload, allocate memory resources from the memory resource pool to a set of processors;

organize data for the given workload in the allocated memory resources according to a data structure, wherein the data structure is a construct implemented to organize the data and is distinct from the data; and

determine whether re-organization of the data in the data structure for the given workload would increase performance of the data structure or conserve memory used by the data structure; and

upon a determination that re-organization would increase performance or conserve memory, take a given action to manage the data structure, wherein the given action detects data similarity or duplicates and, in response thereto, eliminates given data from the data structure.

9. The apparatus as described in claim 8 wherein data similarity or duplicates are detected by:

identifying data blocks that are common to multiple tenants or multiple processes;

upon identifying common data blocks, determining whether the common data blocks are stable according to a configurable threshold;

upon a determination that the common data blocks are stable, eliminating duplicate common data blocks;

associating an instance of the common data blocks as a shared common data block; and

providing the multiple tenants or multiple processes a pointer to the shared common data block.

10. The apparatus as described in claim 9, wherein the computer program instructions are further operative to:

determine that a value for a shared common data block has changed for a first tenant;

determine whether it is computationally less expensive to store a delta value for the shared common data block for the first tenant; and

upon a determination that it is computationally less expensive, store the delta value for the shared common data block for the first tenant.

11. The apparatus as described in claim 10, wherein the computer program instructions are further operative to:

determine whether it is computationally less expensive to store a new value for the shared common data block for the first tenant; and

upon a determination that it is computationally less expensive to store a new value, store the new value for the shared common data block for the first the tenant.

12. The apparatus as described in claim 8, wherein the given action to manage the data comprises:

monitoring data distribution or usage patterns of the data; and

based on the monitoring, applying an optimization to adjust the data structure.

13. The apparatus as described in claim 8, wherein the computer program instructions are further operative to:

profile a data set to produce a data set schema;

periodically back-up the data set without using a processor; and

flush cached data to main memory prior to a backup operation.

14. The apparatus as described in claim 8 wherein the data structure organizes the data based on an access requirement of the given workload.

15. A computer program product in a non-transitory computer readable medium for use in a data processing system to manage memory in a multi-tenant compute environment, the computer program product holding computer program instructions executed in the data processing system and operative to:

organize a memory resource pool;

for a given workload, allocate memory resources from the memory resource pool to a set of processors;

organize data for the given workload in the allocated memory resources according to a data structure, wherein the data structure is a construct implemented to organize the data and is distinct from the data; and

determine whether re-organization of the data in the data structure for the given workload would increase performance of the data structure or conserve memory used by the data structure; and

upon a determination that re-organization would increase performance or conserve memory, take a given action to manage the data structure, wherein the given action detects data similarity or duplicates and, in response thereto, eliminates given data from the data structure.

16. The computer program product as described in claim 15 wherein data similarity or duplicates are detected by:

identifying data blocks that are common to multiple tenants or multiple processes;

upon identifying common data blocks, determining whether the common data blocks are stable according to a configurable threshold;

upon a determination that the common data blocks are stable, eliminating duplicate common data blocks;

associating an instance of the common data blocks as a shared common data block; and

providing the multiple tenants or multiple processes a pointer to the shared common data block.

17. The computer program product as described in claim 16, wherein the computer program instructions are further operative to:

determine that a value for a shared common data block has changed for a first tenant;

determine whether it is computationally less expensive to store a delta value for the shared common data block for the first tenant; and

upon a determination that it is computationally less expensive, store the delta value for the shared common data block for the first tenant.

18. The computer program product as described in claim 17, wherein the computer program instructions are further operative to:

determine whether it is computationally less expensive to store a new value for the shared common data block for the first tenant; and

upon a determination that it is computationally less expensive to store a new value, store the new value for the shared common data block for the first the tenant.

19. The computer program product as described in claim 15, wherein the given action to manage the data comprises:

monitoring data distribution or usage patterns of the data; and

based on the monitoring, applying an optimization to adjust the data structure.

20. The computer program product as described in claim 15, wherein the computer program instructions are further operative to:

profile a data set to produce a data set schema;

periodically back-up the data set without using a processor; and

flush cached data to main memory prior to a backup operation.

21. The computer program product as described in claim 15 wherein the data structure organizes the data based on an access requirement of the given workload.