28-08-2017, 12:29 PM
Despite seemingly endless increases in the amount of storage and ever-lower hardware costs, storage management remains expensive. In addition, users continue to fill larger and larger disks, worsened by the proliferation of large multimedia files and high-speed broadband networks. Storage requirements continue to grow at a rate of 50% per year.
Even worse, the existing hard drive technology is coming to physical limitations, making it more difficult and costly to meet the increasing demands of users. Storage management costs have remained an important component of total storage costs. Even in the 1970s, storage costs at IBM were several times more than hardware costs, and projected to be ten times the cost of hardware. Today, administration costs are five to ten times the cost of the underlying hardware and are actually increasing as a proportion of the cost because administrators have a limited amount of storage each can manage. Up to 47% of storage costs are associated with administrators who manually manipulate files.
Fortunately, significant savings are possible: studies show that more than 20% of all files - representing more than half of the storage - are regenerable. Other studies indicate that 82% -85% of storage is allocated to files that have not been accessed for more than a month. Studies show that storage management has been a problem in the past, remains a problem today, and is only getting worse - all despite increasing disk sizes.
Recent trends have begun to address storage management through virtualization. Morris exposed the idea of Autonomous Computing, which includes "the ability of the system to adjust to its configuration and allocation of resources to achieve predetermined objectives." The elastic quota system is designed to help the management problem through an efficient allocation of storage, while allowing users maximum freedom, all with minimal administrator intervention.
Elastic quotas enter users in an agreement with the system: users can exceed their quota while space is available, under the condition that the system can automatically reclaim the storage when necessary. Users or applications can designate some files as elastic. When space is short, the elastic quota system (Equota) can recover the space of files marked as elastic; Non-elastic archives maintain existing semantics and are counted against persistent user quotas. This report focuses on the elastic space recovery policies and is organized as follows.
Section 2 describes the overall architecture of the policy system. Section 3 discusses the various elastic quota policies. In Section 4 we discuss interesting aspects of elastic quota implementation. Section 5 presents the measures and performance results of various policies. Section 6 discusses the work related to storage space management policies. Finally, Section 7 presents some conclusions and orientations for future work.
Even worse, the existing hard drive technology is coming to physical limitations, making it more difficult and costly to meet the increasing demands of users. Storage management costs have remained an important component of total storage costs. Even in the 1970s, storage costs at IBM were several times more than hardware costs, and projected to be ten times the cost of hardware. Today, administration costs are five to ten times the cost of the underlying hardware and are actually increasing as a proportion of the cost because administrators have a limited amount of storage each can manage. Up to 47% of storage costs are associated with administrators who manually manipulate files.
Fortunately, significant savings are possible: studies show that more than 20% of all files - representing more than half of the storage - are regenerable. Other studies indicate that 82% -85% of storage is allocated to files that have not been accessed for more than a month. Studies show that storage management has been a problem in the past, remains a problem today, and is only getting worse - all despite increasing disk sizes.
Recent trends have begun to address storage management through virtualization. Morris exposed the idea of Autonomous Computing, which includes "the ability of the system to adjust to its configuration and allocation of resources to achieve predetermined objectives." The elastic quota system is designed to help the management problem through an efficient allocation of storage, while allowing users maximum freedom, all with minimal administrator intervention.
Elastic quotas enter users in an agreement with the system: users can exceed their quota while space is available, under the condition that the system can automatically reclaim the storage when necessary. Users or applications can designate some files as elastic. When space is short, the elastic quota system (Equota) can recover the space of files marked as elastic; Non-elastic archives maintain existing semantics and are counted against persistent user quotas. This report focuses on the elastic space recovery policies and is organized as follows.
Section 2 describes the overall architecture of the policy system. Section 3 discusses the various elastic quota policies. In Section 4 we discuss interesting aspects of elastic quota implementation. Section 5 presents the measures and performance results of various policies. Section 6 discusses the work related to storage space management policies. Finally, Section 7 presents some conclusions and orientations for future work.