Ravi Bhagavathula

IP connectivity and DAP

The information technology (IT) revolution, combined with peoples need to access information quickly, has resulted in the explosive growth of the Internet in the past decade. Ubiquitous access to the Internet has become an essential component of a mobile workforce and multiple mechanisms are being devised to ensure seamless connectivity to corporate resources. The authors present a possible use of the available IP connectivity between the airplane and the ground stations for the download of voice/video/data traffic from an airplane onto the ground stations to ease the reliance on blackboxes in a post-incident scenario. A discussion of the simulation test-bed, the results obtained and the practical set of guidelines for their deployment in real-world situations is also included.

Airplane data networks and security issues

The information technology (IT) revolution, combined with peoples need to access information quickly, has resulted in the explosive growth of the Internet in the past decade. Ubiquitous access to the Internet has become an essential component of a mobile workforce and multiple mechanisms are being devised to ensure seamless connectivity to corporate resources. An integrated security framework requires a careful consideration of the security features of the network within an airplane. The passenger network (PN) is used by passengers within the airplane to access network resources on the global Internet. The crew network (CrN), on the other hand, is meant for the crew of the airplane to access resources not only on the global Internet, but also to access resources within the airplanes home network. The control network (CoN) is a strictly regulated network wherein the various components of an airplane interact with each other. As such, only authorized personnel are allowed access to the CoN. In order to facilitate an efficient monitoring of network activity within the PN, the CrN and the CoN, the authors present an in-house network monitoring tool tuned towards the case of a networked airplane that provides real-time warning of impending network threats to allow the network administrators to carry out appropriate responses to intrusions. The network monitoring agents would be located within the individual networks (PN, CrN and CoN) to monitor individual networks. In addition, another sensor would be located within the aircraft access network to ascertain if malicious traffic is introduced into the CrN and/or the CoN.

Timing analysis of block replacement algorithms on disk caches

Cache memories are used to reduce the memory latency in systems. While instruction references of a CPU exhibit high temporal and spatial locality, disk references exhibit very minimal temporal and spatial locality. Owing to the fact that most of the block replacement algorithms exploit the available locality to improve cache performance, they are more effective with CPU instruction caches than with disk caches. This paper presents the results of an investigation of cache write policies and the impact of the Least Recently Used (LRU) and the Segmented LRU (SLRU) block replacement algorithms on the performance of disk caches. To obtain optimal performance at all workloads and cache sizes, an adaptive write caching policy is introduced. The adaptive write caching policy does a dynamic selection of the write policy at run time. Simulations reveal that when the cache size is less than 2 MB, caches employing adaptive write caching policy are 17% faster over caches employing write-back policy. For cache sizes of 16 MB and above the performance improvement is 9%. The performance improvement of caches employing adaptive write caching policy over caches employing write-through policy is 2.65% for cache sizes of 2 MB and is 27%, for cache sizes of 16 MB and above. The adaptive write caching policy yields optimum performance for many of the disk workloads and disk cache sizes.

Mobile IP and virtual private networks

With a growing number of portable computing devices like laptops and personal digital assistants (PDA), the need for seamless connectivity to the global Internet is driving the acceptance of different mobility solutions. Mobile IP is one of the widely accepted mobility solutions for mobile nodes. Though mobile IP caters the seamless connectivity requirements of the mobile nodes, the issues regarding quality of service (QoS) and security remains unsolved. Most of the corporate users like to have seamless connectivity along with security and QoS while they are roaming in the foreign network. The issues related to security and QoS becomes more complicated when dealing with a network on the move (mobile network). A virtual private network (VPN) can be used as an alternative mobility solution to cater the security and QoS in addition to mobility. In the current research work, we evaluate the QoS provided by these alternative mobility solutions as compared to the traditional mobile IP.

Performance of LRU block replacement algorithm with pre-fetching

An economical solution to the need for unlimited amounts of fast memory is a memory hierarchy, which takes advantage of locality and cost/performance of memory technologies. Most of the advanced block replacement algorithms exploit the presence of temporal locality in programs to achieve a better performing cache. A direct fallout of this approach is the increased overhead involved due to the complexity of the algorithm without any significant improvement in the cache performance. The performance of the cache could be improved if spatial locality present in the programs is further exploited. This paper presents the results of the investigation of the impact of pre-fetching techniques on the miss rates due to the basic Least Recently Used (LRU) block replacement algorithm. Simulations reveal an improvement of about 60% in the miss rates for instruction caches due to pre-fetching and a corresponding improvement of about 10% for data caches.

Efficient data storage mechanisms for DAP

The cockpit voice recorder (CVR) and digital flight data recorder (DFDR) are the traditional black boxes used in general and commercial aviation aircrafts. These are used to record vital audio and aircraft parameters. Substantial time and monetary expense are incurred after an aircraft accident to retrieve the black boxes and sometimes the recorders are found damaged and unreadable which further inflates aircraft accident investigation time and expenditures. The CVR typically records the voice conversations within the cockpit on 2 (or 4) different channels for a duration of 30 minutes. The DFDR records the aircrafts vital parameters over the entire duration of a flight. The CVR records information in such a way that only the last 30 minutes of voice is available. As a supplement to the existing CVR/DFDR, the authors present the possible transfer of the acquired voice, video and data from the airplane to the ground stations. This transfer is envisioned to be carried out by (a) utilizing the available data link being employed for IP connectivity between the airplane and the ground station to stream live data, voice and video traffic to the appropriate servers on the ground, or (b) storing the data, voice and video streams locally within the airplane and downloading them to the appropriate servers on the ground station. Since numerous aircraft are expected to be in-flight at any given point of time, the management of the downloaded voice and data within the ground stations could easily become a scalability issue. While file transfer mechanisms like FTP provide considerable flexibility in the deployment of DAP, a scalable means of catering to hundreds of airplanes simultaneously would be the adoption of file I/O and block I/O based data transfer mechanisms. Different I/O mechanisms including (a) network file system (NFS), (b) Internet small computer system interface (iSCSI), and (c) enhanced network block device (ENBD) were considered for the current work.

Port address translation based route optimization for mobile IP

The current mobility management protocol mobile IP (MIP) specifies the use of a home agent (HA) in forwarding datagrams to/from the mobile node (MN). This often leads to the usage of either triangular routing or reverse tunneling. Both these modes of communication introduce additional delay in data transmission between a corresponding node (CN) and a MN, in addition to the wastage of network resources. IETF has proposed certain extensions to MIP to support route optimization. A drawback of the proposed route optimization extensions to MIP is the requirement for the CN to be mobility aware. In this paper, the authors propose a port address translation based route optimization scheme. The proposed route optimization scheme attempts to reduce the overhead and delay involved with traditional mobile communication by means of using port address translation (PAT) and routing the packet using an optimal path. Preliminary analysis carried out by the authors indicates a significant performance improvement compared to that of normal MIP and other route optimization techniques.

Issues with nested mobility

Mobility and Internet access have become an integral part of todays life. The recent advances in computer hardware have also aided in increasing the dependence on Internet connectivity. Protocols like mobile IP were proposed in an effort maintain mobile Internet connections. Even though the primary goal of traditional mobile IP is the support of single host mobility, its extension to support network mobility is an active area of development. Mobile IP needs some additional mobility agents and tunneling to support network mobility. This work explores the possibility of using the mobile IP protocol suite for supporting nested mobility. Issues related to quality of service (QoS), security and scalability within the framework of nested mobility are discussed in this paper. Simulations carried out in Student Routers Lab at Wichita State University reveal deterioration of QoS with nested mobility.

Pre-fetching with the segmented LRU algorithm

An economical solution to the need for unlimited amounts of fast memory is a memory hierarchy, which takes advantage of locality and cost/performance of memory technologies. Most of the advanced block replacement algorithms exploit the presence of temporal locality in programs towards a better performing cache. A direct fallout of this approach is the increased overhead involved due to the complexity of the algorithm without any drastic improvement in the cache performance. The performance of the cache could be improved if spatial locality present in the programs is further exploited. Prefetching of data into the cache even before it is requested is considered to be an effective means of improving the utilization of spatial locality within a stream of references. The Segmented-LRU (S-LRU) algorithm is an advanced block replacement algorithm that reduces the problem of flooding in the cache. This paper investigates the impact of pre-fetching on a cache employing the S-LRU block replacement algorithm.

Pipeline LRU block replacement algorithm

Recent advances in VLSI technology have spurred a tremendous increase in the performance of processors. Due to the slower main memory, there exists a bottleneck in the performance of computer systems. Caching is an effective way to reduce these bottlenecks. With increasing cache sizes, the performance of the processor could be enhanced by using advanced block replacement algorithms like LRU etc. However, due to the presence of the cache in the critical timing path, many processors do not employ these advanced replacement policies. In this paper, the authors present an alternative implementation of block replacement algorithms in CPU caches by modifying the processor pipeline to hide the latency involved in the replacement scheme.