Address label distribution scheduling and multipath cache write-back method for Path ORAM

Abstract

The invention discloses an address label distribution scheduling and multi-path cache write-back method for a Path ORAM. The method comprises four stages of label distribution of new data entering anORAM tree, partial advance and combination during path access, write-back of data in a cache and label distribution of data write-back to the ORAM tree. Label distribution or random distribution basedon memory lines is carried out on the data after the new data enters the memory or the data is written back from the cache; then, the access to the path enters a request queue, and path advance or combination optimization is carried out; access to the whole path is executed, and data of all nodes in the path are sent into a cache; and when the data is returned from the cache, an optimal path is selected for writing back, so that the expected value of the quantity of the overlapped data in the next random path access is maximum. According to the address label distribution scheduling and multi-path cache write-back method for the Path ORAM, the Path ORAM can be rapidly accessed under the condition of very low hardware cost, the execution time is shortened, and the system performance is remarkably improved.

Description

technical field [0001] The present invention relates to the technical field of information security, in particular to a method and device for address label allocation scheduling and multi-path cache write-back for Path ORAM. Background technique [0002] Oblivious RAM (ORAM for short) is a security protocol that protects the memory access mode. This protocol converts the access to a single memory unit into a computationally indistinguishable sequence of accesses with a fixed number of accesses, thus obscuring the original There are access modes. Among them, path-based invisible memory (Path Oblivious RAM, hereinafter referred to as Path ORAM) is currently the most efficient and most feasible ORAM protocol. However, the huge memory access overhead brought by the ORAM protocol will seriously degrade system performance. [0003] In order to reduce the access overhead, the existing technology usually adopts the following methods: first, by identifying and discarding redundant ...

AI Technical Summary

Problems solved by technology

However, the huge memory access overhead brought by the ORAM protocol will seriously reduce system performance

[0003] In order to reduce the access overhead, the existing technology usually adopts the following methods: first, by identifying and discarding redundant request parts between path accesses, thereby reducing a part of the overhead, but this method has certain limitations when there is a dependency between data. Second, according to the phenomenon that the data of the fork Path ORAM will not be read in a short time after being written back, the "read operation" is advanced before the "write operation", although this method can Effectively improve system performance, but the reduced overhead ratio is very small, and the access cost is still high; third, by setting a small ORAM tree as its cache for the main ORAM tree, and this method is similar to the second effect, Path ORAM The memory access overhead is still high

Images