Molecular communication (MC) system is an emerging technology for nanoscale networks. Therefore, there is a requirement to develop a new end-to-end MC model, which may deliver new perceptions into the aspect of these nanoscale networks. This paper aims to implement the MC framework as an endto-end deep reinforcement learning (DRL) auto encoder (AE). The technique enables training of the MC system without any information about the actual channel (medium) model. For training the receiver and transmitter, the proposed techniques are supervised learning and DRL, respectively. The results show that the performance of the DRL autoencoder (AE) based system achieves nearly the same performance as the traditional modulation and demodulation methods in term of bit-error-rate (BER) under the Gaussian noise channel but with less complexity. The proposed technique can also be joint with the other coding methods to improve their performance. INDEX TERMSMolecular communication (MC), deep reinforcement learning (DRL), auto encoder (AE), bit-error-rate (BER).

It is critical to understand the workload characteristics and resource usage patterns of available applications to guide the design and development of architecture of the future large scale servers. In this paper, we analyze the workload performance characteristics of the actual Bank Intermediary Business (BIB) characteristics with BIBmodel and BIBbench design work, and propose BIB performance workload and use case definitions. The analysis and comparisons of workload and use case illustrate that the workload performance characteristics of BIB is totally different with TPC benchmarks. With the development of economy and technology, the requirements for BIB servers are important in modeling, benchmarks developing and workload performance characteristics studying are increased nowadays.

Importance measures are widely used to characterize the contribution of components to the system performance such as reliability, availability, risk, etc, and thus give great help in identifying system weaknesses and prioritizing system improvement activities. Although much work has been carried out on component importance analysis, most studies only concern the consistent states of components within which components exhibit consistent performance until state changes happen. Unfortunately, field data shows that many transient faults in components may result in severe consequences without causing any state changes, and, this can lead to a misunderstanding of component importance. This paper focuses on the reliability importance analysis in presence of transient faults, and proposes a composite measure for evaluation. A sample series parallel system is analyzed to illustrate the use of this measure.

Availability is one of the key features for transaction processing systems. Fault injection is one of the important techniques to hasten availability tests. A fault injection platform is designed in this paper for IA64 system which is being widely used in transaction processing business. The fault injection platform is implemented using client/server mode and a series of fault injection tools are accomplished which covers CPU faults, memory faults, disk faults, IO faults, file system faults and system call boundary errors. A group of experiments on a typical IA64 server are described and performed, and the experiment results validate the effectiveness of the fault injection platform.