WorldCat Identities

Malladi, Krishna Teja

Works: 3 works in 3 publications in 1 language and 3 library holdings
Genres: Academic theses 
Roles: Author
Publication Timeline
Most widely held works by Krishna Teja Malladi
Development of a decision support tool for optimizing the short-term logistics of forest-based biomass( )

1 edition published in 2018 in English and held by 1 WorldCat member library worldwide

Highlights: Transshipment and routing models are developed for short-term biomass logistics. The models include biomass storage, pre-processing, and truck routing decisions. Models are applied to a large forest-based biomass logistics company. Average reduction of 12% in cost and fuel consumption is observed using the models. An Excel-based decision support tool is developed for the company. Abstract: High cost of logistics is one of the barriers of using forest-based biomass for energy and fuel production. Biomass logistics is complex and includes interdependent decisions related to storage, pre-processing and transportation. While these decisions have been considered in numerous medium-term planning models, those for short-term planning are limited. The existing models focused only on optimal truck routing without considering intermediate storage facilities which are essential to match biomass supply and demand. In this study, a decomposition-based approach is used and optimization models are developed for the short-term planning of a large biomass logistics company located in the Lower Mainland region of British Columbia, Canada. The company deals with collection, storage, pre-processing and transportation of biomass. Several operational constraints related to truck-location compatibilities and truck-biomass compatibilities arising from heterogeneity of trucks and biomass types which further complicate the logistics planning are incorporated in the models. First, a transshipment model is developed and solved using a mixed integer formulation to determine comminution schedules and the number of truckloads of each biomass type to be transported each day using each type of truck. Then, a routing model, which uses the results of the transshipment model, is developed to determine the optimal routing for the available trucks. A decision support tool to optimize the company's weekly transportation and comminution operations is also developed for the company. Experiments were conducted on real data from the company over a span of four weeks. The results indicate 12% reduction in the total average cost and a similar reduction in fuel consumption compared to the actual routes implemented by the company. It is suggested that savings could be obtained by using larger trucks for longer distance transportation and smaller trucks for shorter distances. Direct delivery of biomass from suppliers to customers, bypassing the yard, could result in cost savings
Energy proportional memory systems by Krishna Teja Malladi( )

1 edition published in 2013 in English and held by 1 WorldCat member library worldwide

The horizon of computer systems is changing. Moore's law is power constrained, making energy efficiency the primary goal of computer systems. In particular, memory energy has emerged as the key efficiency bottleneck, owing to extensive research in low power CPU architectures. On the other hand, the popularity of modern day internet services has lead to a vast expansion of datacenters, magnifying the importance of efficient computer system design. With datacenter energy provisioning and running costs running into millions of dollars, energy is an even more important criterion in datacenters. To increase computer system energy efficiency, we need memory systems that keep pace with processor efficiency gains. Currently, servers use DDR3 memory, which is designed for high bandwidth but not for energy proportionality. A system using 20% of the peak DDR3 bandwidth consumes 2.3x the energy per bit compared to the energy consumed by a system with fully utilized memory bandwidth. Nevertheless, many datacenter applications stress memory capacity and latency but not memory bandwidth. In response, we architect server memory systems using mobile DRAM devices i.e. LPDDR2, trading peak bandwidth for lower energy consumption per bit and more efficient idle modes. We demonstrate 3-5x lower memory power, better proportionality, and negligible performance penalties for datacenter workloads. Noting that LPDDR2 has performance impact for high bandwidth applications, we explore low-power, high-speed interfaces that can be applied to a broad variety of applications. We re-think DRAM power modes by modeling and characterizing inter-arrival times for memory requests to determine the properties an ideal power mode should have. This analysis indicates that even the most responsive of today's power modes are rarely used. As a result, up to 88% of memory is spent idling in an active mode. This analysis indicates that power modes must have much shorter exit latencies than they have today. Wake-up latencies less than 100ns are ideal. To address these challenges, we present MemBlaze, an architecture with DRAMs and links that are capable of fast powerup, which provides more opportunities to powerdown memories. By eliminating DRAM chip timing circuitry, a key contributor to powerup latency, and by shifting timing responsibility to the controller, MemBlaze permits data transfers immediately after wake-up and reduces energy per transfer by 50% with no performance impact. Alternatively, in scenarios where DRAM timing circuitry must remain, we explore mechanisms to accommodate DRAMs that powerup with less than perfect interface timing. We present MemCorrect which detects timing errors while MemDrowsy lowers transfer rates and widens sampling margins to accommodate timing uncertainty in situations where the interface circuitry must recalibrate after exit from powerdown state. Combined, MemCorrect and MemDrowsy still reduce energy per transfer by 50% but incur modest (e.g., 10%) performance penalties. Finally, we demonstrate that we need to re-evaluate our design choices for last level caches whose static power is beginning to compete with the dynamic energy of new and efficient memory systems like LPDDR2. We propose a novel metric called "Average Memory Access Energy" that quantifies energy costs of the memory hierarchy and helps us make efficient design choices. We demonstrate how the low energy memory hierarchy helps us in lowering system operating costs
Cluster Restricted Maximum Weight Clique Problem and Linkages with Satellite Image Acquisition Scheduling by Krishna Teja Malladi( )

1 edition published in 2014 in English and held by 1 WorldCat member library worldwide

We consider the cluster-restricted maximum weight clique problem (CRCP), a variation of the well-known maximum weight clique problem. While CRCP itself is mathematically interesting, our motivation to study the problem primarily comes from its application in the area of Satellite Image Acquisition Scheduling. Earth observing satellites revolve around the earth in specific orbits and takes images of prescribed areas requested by the clients. Not every region can be fully acquired in a single satellite pass. This necessitates the division of the region into multiple strips. There might be several image acquisition opportunities for each strip as the satellites have agility in their rolling angles. Then the Satellite Image Acquisition Scheduling Problem (SIASP) is to select the opportunities to acquire as many images as possible, without repetition, within a planning horizon while considering the image priorities and energy constraints. SIASP has a piecewise linear objective function to favor the completion of an image acquisition request and try to avoid partial acquisition of many requests. Extensive experimental study is provided on randomly generated instances based on the predicted statistics given by MDA, Richmond, Canada. These experiments are intended as a preliminary investigation on image acquisition scheduling for the Canadian RADARSAT Constellation Mission (RCM), a constellation of three satellites, which is planned to be launched in 2018.SIASP can be modeled as a CRCP with piecewise linear objective function. We provide integer programming (IP) formulations for CRCP with linear and piecewise linear objective function. We also suggest heuristic (metaheuristic) algorithms that exploit the power of modern IP solvers such as CPLEX. Experimental results using the heuristic algorithms on DIMACS and BOSHLIB benchmark instances for the clique problem are reported. Finally, an exact algorithm for CRCP along with some theoretical analysis is presented
Audience Level
Audience Level
  General Special  
Audience level: 0.00 (from 0.00 for Developmen ... to 0.00 for Developmen ...)

Alternative Names
Malladi, K. T.