Dependability analysis for characterizing Google cluster reliability

Cloud solutions are emerging as a new suitable way of transforming traditional IT data centers to highly available and reliable computing resources for hosting critical applications and data. However, software and hardware failures are a common problem in cloud datacenters that can lead to harmful damages. In this paper, we analyze the physical server failures in the Google cloud datacenter. We study the Google cluster properties to investigate the relationship among physical servers' failure rate and jobs failure events. The failure rate of Google cluster executed jobs and servers is taken into consideration during a 29‐day period. We present a reliability model for Google cluster physical machines using the continuous time Markov chains according to this observation. We attempt to analyze the obtained model through SHARPE software packages to improve the understanding of failure events in the Google cloud cluster. We also explore the cluster availability based on parameters like steady‐state availability, steady‐state unavailability, mean time to failure, and mean time to repair in the Google cluster.     

Weighted image fusion using cross bilateral filter and non-subsampled contourlet transform

Multidimensional Systems and Signal Processing - This paper provides a fusion technique for multi-focus imaging using cross bilateral filter and non-subsampled contourlet transform. The snapshots...     

CCGA packages for space applications

Commercial-off-the-shelf (COTS) area array packaging technologies in high reliability versions are now being considered for applications, including use in a number of NASA electronic systems being utilized for both the Space Shuttle and Mars Rover missions. Indeed, recently a ceramic package version specifically tailored for high reliability applications was used to provide the processing power required for the Spirit and Opportunity Mars Rovers built by NASA-JPL. Both Rovers successfully completed their 3-months mission requirements and continued exploring the Martian surface for many more moths, providing amazing new information on previous environmental conditions of Mars and strong evidence that water exists on Mars.Understanding process, reliability, and quality assurance (QA) indicators for reliability are important for low risk insertion of these newly available packages in high reliability applications. In a previous investigation, thermal cycle test results for a non-functional daisy-chained peripheral ceramic column grid array (CCGA) and its plastic ball grid array (PBGA) version, both having 560 I/Os, were gathered and are presented here. Test results included environmental data for three different thermal cycle regimes (−55/125 °C, −55/100 °C, and −50/75 °C). Detailed information on these—especially failure type for assemblies with high and low solder volumes—are presented. The thermal cycle test procedure followed those recommended by IPC-9701 for tin–lead solder joint assemblies. Its revision A covers guideline thermal cycle requirements for Pb-free solder joints. Key points on this specification are also discussed.In a recent investigation a fully populated CCGA with 717 I/Os was considered for assembly reliability evaluation. The functional package is a field-programmable gate array that has much higher processing power than its previous version. This new package is smaller in dimension, has no interposer, and has a thinner column wrapped with copper for reliability improvement. This paper will also present thermal cycle test results for assemblies of this and its plastic package version with 728 I/Os, both of which were exposed to four different cycle regimes. Two of these cycle profiles are specified by IPC-9701A for tin–lead, namely, −55 to 100 °C and −55 to 125 °C. One is a cycle profile specified by Mil-Std-883, namely, −65/150 °C, generally used for ceramic hybrid packages screening and qualification. The last cycle is in the range of −120 to 85 °C, a representative of electronic systems directly exposed to the Martian environment without use in a thermal control enclosure. Per IPC-9701A, test vehicles were built using daisy chain packages and were continuously monitored and/or manually checked for opens at intervals. The effects of many process and assembly variables—including corner staking commonly used for improving resistance to mechanical loading such as drop and vibration loads—were also considered as part of the test matrix. Optical photomicrographs were taken at various thermal cycle intervals to document damage progress and behavior. Representative samples of these are presented along with cross-sectional photomicrographs at higher magnification taken by scanning electron microscopy (SEM) to determine crack propagation and failure analyses for packages.     

An Intra-domain Mobility Handling Scheme Across All-IP Wireless Networks

The rapid growth of wireless network technology such as HSDPA and WiMAX, has lead to greater demand for access to Internet via mobile hosts. Supporting mobile connection with fast and smooth roaming across heterogeneous wireless technologies has been an important challenge over past years. In this paper, a novel multilayer scheme for QoS-aware intra-domain mobility management is proposed. The mobility support capability is embedded in key components for the domain access network, namely, the Paging Access Routers and the Mobility-support Anchor Servers (MASs). The MASs are organized in three layers; starting from the top layer Superior-MASs, Middle-MASs and Inferior-MASs, respectively. The proposed scheme identified mobility support functionality, includes intra-domain anchor specification, route optimization algorithm, intra/inter-anchor mobility support, paging and authentication management. Simulation results of the proposed scheme show fair performance especially in the presence of QoS sensitive services.     

Performance of Layered Steered Space–Time Codes in Wireless Systems

Layered Steered Space–Time Codes (LSSTC) is a recently proposed multiple-input multiple-output (MIMO) system that combines the benefits of vertical Bell Labs space–time (VBLAST) scheme, space–time block codes (STBC) and beamforming. In this paper, we derive the error performance and capacity of a single-user LSSTC system. The analysis is general enough to any layer ordering and modulation schemes used. In addition, the derived analysis is general for any LSSTC structure in which layers may have different number of antenna arrays and may be assigned power according to any power allocation. Furthermore, we analytically investigate the tradeoff between the main parameters of the LSSTC system, i.e., diversity, multiplexing and beamforming. Our results give recursive expressions for the probability of error for LSSTC which showed nearly perfect match to the simulation results. Results have also revealed the possibility of designing an adaptive system in which it was shown that combining beamforming, STBC, and VBLAST has better performance than VBLAST at high SNR range.     

Minimum Autocorrelation Spreading Codes

Intersymbol Interference (ISI) always appears when the communication channel is a multi-path channel. Many methods are used to reduce the effect of ISI. In this paper we focus on the autocorrelation property of the spreading codes and the role it plays in minimizing the ISI effect, where we obtain the spreading codes with minimum autocorrelation property. The results of comparing the average autocorrelation of the obtained codes with that of the well known Hadamard codes show a great enhancement in the performance, where for example, for codes of length 8 a gain of 408% was achieved at one shift, and for codes of length 16 a gain of 530% was achieved at one shift. Ahmad I. Amayrah was born in Amman in 1976. He received his Master of Science in Electrical Engineering from University of Jordan in 2003. He works as a lecturer in Al-Balqaa' University Abdallah K. Farraj was born in Amman in 1977. He earned his Bachelor of Science and Master of Science degrees in Electrical Engineering from University of Jordan, Amman, Jordan. He is currently working towards a PhD degree at the Department of Electrical and Computer Engineering, Texas A&M University. He is a fellow of the Graduate Teaching Academy, Texas A&M University, and he was awarded the Fulbright Scholarship for the 2005–2006 and 2006–2007 academic years.     

In Situ Liquid-Cell TEM Observation of Multiphase Classical and Nonclassical Nucleation of Calcium Oxalate

Calcium oxalate (CaOx) is the major phase in kidney stones and the primary calcium storage medium in plants. CaOx can form crystals with different lattice types, water contents, and crystal structures. However, the conditions and mechanisms leading to nucleation of particular CaOx crystals are unclear. Here, liquid-cell transmission electron microscopy and atomistic molecular dynamics simulations are used to study in situ CaOx nucleation at different conditions. The observations reveal that rhombohedral CaOx monohydrate (COM) can nucleate via a classical pathway, while square COM can nucleate via a non-classical multiphase pathway. Citrate, a kidney stone inhibitor, increases the solubility of calcium by forming calcium-citrate complexes and blocks oxalate ions from approaching calcium. The presence of multiple hydrated ionic species draws additional water molecules into nucleating CaOx dihydrate crystals. These findings reveal that by controlling the nucleation pathways one can determine the macroscale crystal structure, hydration state, and morphology of CaOx.     

Differential current conveyor based current comparator

A New differential current conveyor based current comparator is presented in this paper. Differential current conveyor II (DCCII) is designed, modified, and exploited as a comparator with reduced propagation delay and power consumption. New DCCII decreases propagation delay and increases comparator accuracy considerably. Simulation results using Hspice and 0.18 μm CMOS technology with 1.8V supply voltage confirms a less than 0.63 ns propagation delay at ±1 μA input current. Average power dissipation in ±1 μA input current has a value of 300 μW.     

Public Key Cryptosystem Based on Low Density Lattice Codes

McEliece and Goldreich–Goldwasser–Halevi (GGH) cryptosystems are two instances of code and lattice-based cryptosystems whose security are based on the hardness of coding theoretic and lattice problems, respectively. However, such cryptosystems have a number of drawbacks which make them inefficient in practice. On the other hand, low density lattice codes (LDLCs) are practical lattice codes which can achieve capacity over additive white Gaussian noise channel and also can be encoded and decoded efficiently. This paper introduces a public key cryptosystem based on Latin square LDLCs, by which a relationship can be attained between code and lattice-based cryptography. In this way, we can exploit the efficient properties of codes and lattices, simultaneously to improve the security and efficiency of the proposed scheme. For instance, the security of this scheme is based on the hard problems related to lattices, i.e., closest vector problem and shortest basis problem, which in turn lead to increase the security level. On the other hand, we exploit the low complexity decoding algorithm of LDLCs to reduce the computational complexity. Moreover, this property allows using the larger values of the codeword length. Also, we use the special Gaussian vector, whose variance is upper bounded by Poltyrev bound, as the perturbation (error) vector. These strategies make the proposed scheme to be secure against the conventional cryptanalytic attacks.