Buffer Cache Level Encryption for Embedded Secure Operating System

Due to the constantly growing need for security and efficiencyin accessing stored data, there is also constant search for betterstorage media that bring this warranty. Previous research has led tothe creation of libraries for parallel processing on GPU to performspeculative encryption of data. The efficiency of this method wasdemonstrated in its use in cryptographic file systems. However, theapplicability and effectiveness of these methods using only CPUs remained an open problem. In order to test the feasibility of specu-lative cryptography in computing environments with CPUs only, this work presents a library version for speculative encryption withparallel processing on multicore processors and its adaptation to the EncFS++ file system in user space. The tests carried out show con-siderable increases in throughput in the most varied environments, both physical and virtual machines, using CPUs that may or maynot support the AES-NI instruction set for encryption acceleration.

Virtual memory was developed to automate the movement of program code and data between main memory and secondary storage to give the appearance of a single. This technique greatly simplified the programmer’s job, particularly when program code and data exceeded the main memory’s size. In modern operating system, secure file system can protect confidential data from unauthorized access. However, once an authorized process has accessed data from a cryptographic file system, the data can appear as plaintext in the unprotected virtual memory backing store, even after system shutdown. This paper proposes new approach for adding multilevel security capability to virtual memory management. It uses partial swap encryption based on process’s security policy. Volatile encryption keys are chosen randomly, and remain valid only for short time periods. Performance comparison shows the total overhead of proposed method.

Cryptographic file systems ensure information confidentiality even in the presence of attacks which successfully circumvented the access controls of the underlying operating system. The paper presents our new system called GSFS which overcomes several severe deficiencies found in popular cryptographic file systems. In contrast to these systems GSFS offers a transparent group-based access control which supports a comfortable sharing of file between user groups. A prototype is available running integrated into the Linux operating system.KeywordsFile SystemCryptographyAccess ControlManagementOperating System

This paper proposes a backup system based on mirroring filesystem GMFS has been developed to mirror data in realtime on the filesystem layer. The GMFS is a stackable filesystem which flexibly mirrors without changing the existing environment by operating as a wrapper of other filesystems.Because the conventional mirroring technology utilizes the mirroring function on the device layer or needs a special filesystem, the allocation of the disk and the specific format of the filesystem are needed, and so the disk design is fixed. Therefore, the conventional mirroring technology cannot adjust when the mirroring function not assumed will be needed later. In this situation, the mechanism that adds the mirroring function without changing the existing disk design is necessary. The GMFS conducts the operation of other filesystems transparently, thereby users need not be aware of the GMFS. The conventional filesystem looks as if it performs mirroring data by itself. GMFS can therefore add the function that mirrors in realtime without destroying the existing environment.GMFS uses NFS which is a typical network file system to communicate with an existing environment. The throughput of reading and writing has been improved by adopting the method to call system function of NFS from the inside of the filesystem.We developed this filesystem, and evaluated the performance from the viewpoint of throughput and system call speed and CPU loads. As a result, it was shown that there was no problem in the viewpoint of the performance compared with the conventional filesystem, and the throughput of the read and write of GMFS was 2.0 times faster than conventional mirroring filesystem.

Distributed file system is the basic platform for big data analysis. For VM, database, log generation and other file processing, the traditional distributed file system for big data analysis cannot meet the IOPS requirements. Flash storage is becoming more common and costs are slowly coming down but it is still not practical to store a large quantity of your data on flash. Making the best use of flash storage requires automation. The storage needs to be smart enough to handle the movement of data between flash and slower storage so applications don't need to understand the underlying architecture. In addition to smart storage sometimes it makes the most sense to allow an application to provide hints or direction to the storage so data can be place most efficiently. This paper presents a method for constructing a set of highly available write cache and Local read-only buffer using SSD. By configuring recovery log and meta-data separation management, the method collaborates with the cache and reduces the read-write delay of massive small files to meet the requirement of high IOPS application.

This chapter is all about the Windows Embedded Compact 7 registry and how it relates to device drivers. A general discussion about the registry and its role in the system is followed by a thorough discussion of all device driver related registry topics, including how the OS loads device drivers; naming conventions of device drivers, so that applications can interact with device drivers via the file system; which keys are required to set by the developer of a device driver and which keys are optional; and how to setup registry entries for user mode device drivers, and how to put it all together in practical terms.

Flash memory has become virtually indispensable in most mobile devices. In order for mobile devices to operate successfully, it is essential that the flash memory be controlled correctly through the device driver software. However, as is typical for embedded software, conventional testing methods often fail to detect hidden flaws in the complex device driver software. This deficiency incurs significant development and operation overheads to the manufacturers. Model checking techniques have been proposed to compensate for the weaknesses of conventional testing methods through exhaustive analyses. These techniques, however, require significant manual efforts to create an abstract target model and, thus, are not widely applied in industry. In this project, we applied a model checking technique based on a Boolean satisfiability (SAT) solver. One advantage of SAT-based model checking is that a target C code can be analyzed directly without an abstract model, thereby enabling automated and bit-level accurate verification. In this project, we have applied CBMC, a SAT-based software model checker, to the unit testing of the Samsung OneNANDtrade device driver. Through this project, we detected several bugs that had not been discovered previously.

In 1987, Gray and Putzolo presented the five-minute rule, which was reviewed and renewed ten years later in 1997. With the advent of flash memory in the gap between traditional RAM main memory and traditional systems, the five-minute rule now applies to large pages appropriate for today's disks and their fast transfer bandwidths, and it also applies to flash disks holding small pages appropriate for their fast access latency.Flash memory fills the gap between RAM and disks in terms of many metrics: acquisition cost, access latency, transfer bandwidth, spatial density, and power consumption. Thus, within a few years, flash memory will likely be used heavily in operating systems, file systems, and database systems. Research into appropriate system architectures is urgently needed.The basic software architectures for exploiting flash in these systems are called extended buffer pool and extended disk here. Based on the characteristics of these software architectures, an argument is presented why operating systems and file systems on one hand and database systems on the other hand will best benefit from flash memory by employing different software architectures.

In most operating systems which are customized for disk-based storage system, the replacement algorithm concerns only the number of memory hits. However, flash memory has different read and write cost in the aspects of time and energy so the replacement algorithm with flash memory should consider not only the hit count but also the replacement cost caused by selecting dirty victims. The replacement cost of dirty page is higher than that of clean page with regard to both access time and energy consumption. In this paper, we propose the Clean-First LRU (CFLRU) replacement algorithm that exploits the characteristics of flash memory. CFLRU splits the LRU list into the working region and the clean-first region and adopts a policy that evicts clean pages preferentially in the clean-first region until the number of page hits in the working region is preserved in a suitable level. Using the trace-driven simulation, the proposed algorithm reduces the average replacement cost by 28.4% in swap system and by 26.2% in buffer cache, compared with LRU algorithm. We also implement the CFLRU algorithm in the Linux kernel and present some optimization issues.

While flash memory has been widely adopted for storage systems for various embedded systems, issues on performance and reliability have started receiving growing attention in recent years. How to provide efficient roll back and quick mounting for flash-memory file systems has become important research topics in recent years, in addition to the work on effective garbage collection and superb run-time performance. Such an observation motivates our work on the investigation of efficient initialization and crash recovery of flash-memory file systems based on log structures. A methodology is proposed for the acceleration of mounting and crash recovery for log-based file systems. A system prototype based on a well-known flash-memory file system YAFFS was implemented with performance evaluation. The experimental results show that the proposed methodology can reduce the mounting time significantly, regardless of whether the file system is properly unmounted.

Among the different existing cryptographic file systems, EncFS has a unique feature that makes it attractive for backup setups involving untrusted (cloud) storage. It is a file-based overlay file system in normal operation (i.e., it maintains a directory hierarchy by storing encrypted representations of files and folders in a specific source folder), but its reverse mode allows to reverse this process: Users can mount deterministic, encrypted views of their local, unencrypted files on the fly, allowing synchronization to untrusted storage using standard tools like rsync without having to store encrypted representations on the local hard drive. So far, EncFS is a single-user solution: All files of a folder are encrypted using the same, static key; file access rights are passed through to the encrypted representation, but not otherwise considered. In this paper, we work out how multi-user support can be integrated into EncFS and its reverse mode in particular. We present an extension that a) stores individual files' owner/group information and permissions in a confidential and authenticated manner, and b) cryptographically enforces thereby specified read rights. For this, we introduce user-specific keys and an appropriate, automatic key management. Given a user's key and a complete encrypted source directory, the extension allows access to exactly those files the user is authorized for according to the corresponding owner/group/permissions information. Just like EncFS, our extension depends only on symmetric cryptographic primitives.

Stackable Cryptographic File Systems have been adopted as one of the dominant solutions to transparently improve data security. However, how to reduce the performance degradation induced by encryption-operations is still an open problem. To solve the performance degradation problem, Load-aware Adaptive Cache Management (LACM) scheme is proposed, which utilizes a load-aware dynamic scheduling strategy to improve cache efficiency from two aspects: load-aware redundancy elimination and non-redundant cache conversion. Experimental results show that the proposed solution can provide 20.03%~34.39% latency decrease, and supply 12.82%~34.77% performance improvement for application-level workloads. LACM can also be easily adapted to other stackable cryptographic/compressed file systems, such as NCryptfs, stackable compress file system and etc.

In embedded systems, it brings great benefits to use NAND flash for dynamic data exchange like using RAM, since RAM is usually expensive and consumes much energy. However, the major hardware limitations of NAND flash make it difficult to directly and dynamically read\write\erase on NAND flash. Traditional virtual memory systems are all based on flash translation layer and file system, which lead to frequently flash read\write\erase and more RAM requirement. To address this problem, we propose a Comet Circle Model (CCM) to emulate RAM for low cost embedded systems in this paper. CCM dynamically makes the distribution of valid pages in NAND flash follow a fixed function through integrating its physical page-level mapping and circularly write-back techniques. As a result, CCM can select a block without valid pages to write data almost every time in data writing, which avoids data moving among blocks. Both theoretical analysis and experimental results show that the data moving times and erase counts of CCM closely reach the theoretical lower bounds.

The buffer cache plays an essential role in smoothing the gap between the upper level computational components and the lower level storage devices. A good buffer cache management scheme should be beneficial to not only the computational components, but also the storage components by reducing disk I/Os. Existing cache replacement algorithms are well optimized for disks in normal mode, but inefficient under faulty scenarios, such as a parity-based disk array with faulty disk(s). To address this issue, we propose a novel penalty-aware buffer cache replacement strategy, named Victim Disk(s) First (VDF) cache, to improve the reliability and performance of a storage system consisting of a buffer cache and disk arrays. VDF cache gives higher priority to cache the blocks on the faulty disks when the disk array fails, thus reducing the I/Os addressed directly to the faulty disks. To verify the effectiveness of the VDF cache, we have integrated VDF into the popular cache algorithms least frequently used (LFU) and least recently used (LRU), named VDF-LFU and VDF-LRU, respectively. We have conducted intensive simulations as well as a prototype implementation for disk arrays to tolerate one disk failure (RAID-5) and two disk failures (RAID-6). The simulation results have shown that VDF-LFU can reduce disk I/Os to surviving disks by up to 42.3 percent in RAID-5 and 50.7 percent in RAID-6, and VDF-LRU can reduce those by up to 36.2 percent in RAID-5 and 48.9 percent in RAID-6. Our measurement results also show that VDF-LFU can speed up the online recovery by up to 46.3 percent in RAID-5 and 47.2 percent in RAID-6 under spare-rebuilding mode, or improve the maximum system service rate by up to 47.7 percent in RAID-5 under degraded mode without a reconstruction workload. Similarly, VDF-LRU can speed up the online recovery by up to 34.6 percent in RAID-5 and 38.2 percent in RAID-6, or improve the system service rate by up to 28.4 percent in RAID-5.

ZTL: A block layer ZNS driver