AdFT: An Adaptive Framework for Fault Tolerance on Large Scale Systems using Application Malleability

Exascale systems of the future are predicted to have mean time between failures (MTBF) of less than one hour. Malleable applications, where the number of processors on which the applications execute can be changed during executions, can make use of their malleability to better tolerate high failure rates. We present AdFT, an adaptive fault tolerance framework for long running malleable applications to maximize application performance in the presence of failures. AdFT framework includes cost models for evaluating the benefits of various fault tolerance actions including checkpointing, live-migration and rescheduling, and runtime decisions for dynamically selecting the fault tolerance actions at different points of application execution to maximize performance. Simulations with real and synthetic failure traces show that our approach outperforms existing fault tolerance mechanisms for malleable applications yielding up to 23% improvement in application performance, and is effective even for petascale systems and beyond.

Frequency domain turbo equalization for MIMO-CPSC systems with large delay spreads

In this paper, we consider low-complexity turbo equalization for multiple-input multiple-output (MIMO) cyclic prefixed single carrier (CPSC) systems in MIMO inter-symbol interference (ISI) channels characterized by large delay spreads. A low-complexity graph based equalization is carried out in the frequency domain. Because of the reduction in correlation among the noise samples that happens for large frame sizes and delay spreads in frequency domain processing, improved performance compared to time domain processing is shown to be achieved. This improved performance is attractive for equalization in severely delay spread ISI channels like ultrawideband channels and underwater acoustic channels.

Factor graph based joint detection/decoding for LDPC coded large-MIMO systems

In this paper, we employ message passing algorithms over graphical models to jointly detect and decode symbols transmitted over large multiple-input multiple-output (MIMO) channels with low density parity check (LDPC) coded bits. We adopt a factor graph based technique to integrate the detection and decoding operations. A Gaussian approximation of spatial interference is used for detection. This serves as a low complexity joint detection/decoding approach for large dimensional MIMO systems coded with LDPC codes of large block lengths. This joint processing achieves significantly better performance than the individual detection and decoding scheme.

EXIT Chart Based Design of Irregular LDPC Codes for Large-MIMO Systems

In this letter, we characterize the extrinsic information transfer (EXIT) behavior of a factor graph based message passing algorithm for detection in large multiple-input multiple-output (MIMO) systems with tens to hundreds of antennas. The EXIT curves of a joint detection-decoding receiver are obtained for low density parity check (LDPC) codes of given degree distributions. From the obtained EXIT curves, an optimization of the LDPC code degree profiles is carried out to design irregular LDPC codes matched to the large-MIMO channel and joint message passing receiver. With low complexity joint detection-decoding, these codes are shown to perform better than off-the-shelf irregular codes in the literature by about 1 to 1.5 dB at a coded BER of 10(-5) in 16 x 16, 64 x 64 and 256 x 256 MIMO systems.

Multiuser detection in large-dimension multicode MIMO-CDMA systems with higher-order modulation

In this paper, we are interested in high spectral efficiency multicode CDMA systems with large number of users employing single/multiple transmit antennas and higher-order modulation. In particular, we consider a local neighborhood search based multiuser detection algorithm which offers very good performance and complexity, suited for systems with large number of users employing M-QAM/M-PSK. We apply the algorithm on the chip matched filter output vector. We demonstrate near-single user (SU) performance of the algorithm in CDMA systems with large number of users using 4-QAM/16-QAM/64-QAM/8-PSK on AWGN, frequency-flat, and frequency-selective fading channels. We further show that the algorithm performs very well in multicode multiple-input multiple-output (MIMO) CDMA systems as well, outperforming other linear detectors and interference cancelers reported in the literature for such systems. The per-symbol complexity of the search algorithm is O(K2n2tn2cM), K: number of users, nt: number of transmit antennas at each user, nc: number of spreading codes multiplexed on each transmit antenna, M: modulation alphabet size, making the algorithm attractive for multiuser detection in large-dimension multicode MIMO-CDMA systems with M-QAM.

Cooperative Particle Swarm Optimization Based Receiver for Large-Dimension MIMO-ZPSC Systems

In this paper, we propose a cooperative particle swarm optimization (CPSO) based channel estimation/equalization scheme for multiple-input multiple-output zero-padded single-carrier (MIMO-ZPSC) systems with large dimensions in frequency selective channels. We estimate the channel state information at the receiver in time domain using a PSO based algorithm during training phase. Using the estimated channel, we perform information symbol detection in the frequency domain using FFT based processing. For this detection, we use a low complexity OLA (OverLap Add) likelihood ascent search equalizer which uses minimum mean square (MMSE) equalizer solution as the initial solution. Multiple iterations between channel estimation and data detection are carried out which significantly improves the mean square error and bit error rate performance of the receiver.

Low-complexity near-optimal signal detection in underdetermined large-MIMO systems

In this paper, we consider signal detection in nt × nr underdetermined MIMO (UD-MIMO) systems, where i) nt >; nr with a overload factor α = nt over nr >; 1, ii) nt symbols are transmitted per channel use through spatial multiplexing, and iii) nt, nr are large (in the range of tens). A low-complexity detection algorithm based on reactive tabu search is considered. A variable threshold based stopping criterion is proposed which offers near-optimal performance in large UD-MIMO systems at low complexities. A lower bound on the maximum likelihood (ML) bit error performance of large UD-MIMO systems is also obtained for comparison. The proposed algorithm is shown to achieve BER performance close to the ML lower bound within 0.6 dB at an uncoded BER of 10-2 in 16 × 8 V-BLAST UD-MIMO system with 4-QAM (32 bps/Hz). Similar near-ML performance results are shown for 32 × 16, 32 × 24 V-BLAST UD-MIMO with 4-QAM/16-QAM as well. A performance and complexity comparison between the proposed algorithm and the λ-generalized sphere decoder (λ-GSD) algorithm for UD-MIMO shows that the proposed algorithm achieves almost the same performance of λ-GSD but at a significantly lesser complexity.

Lattice Reduction Aided Detection in Large-MIMO Systems

Lattice reduction (LR) aided detection algorithms are known to achieve the same diversity order as that of maximum-likelihood (ML) detection at low complexity. However, they suffer SNR loss compared to ML performance. The SNR loss is mainly due to imperfect orthogonalization and imperfect nearest neighbor quantization. In this paper, we propose an improved LR-aided (ILR) detection algorithm, where we specifically target to reduce the effects of both imperfect orthogonalization and imperfect nearest neighbor quantization. The proposed ILR detection algorithm is shown to achieve near-ML performance in large-MIMO systems and outperform other LR-aided detection algorithms in the literature. Specifically, the SNR loss incurred by the proposed ILR algorithm compared to ML performance is just 0.1 dB for 4-QAM and < 0.5 dB for 16-QAM in 16 x 16 V-BLAST MIMO system. This performance is superior compared to those of other LR-aided detection algorithms, whose SNR losses are in the 2 dB to 9 dB range.

Channel Hardening-Exploiting Message Passing (CHEMP) Receiver in Large-Scale MIMO Systems

In this paper, we propose a multiple-input multiple-output (MIMO) receiver algorithm that exploits channel hardening that occurs in large MIMO channels. Channel hardening refers to the phenomenon where the off-diagonal terms of the matrix become increasingly weaker compared to the diagonal terms as the size of the channel gain matrix increases. Specifically, we propose a message passing detection (MPD) algorithm which works with the real-valued matched filtered received vector (whose signal term becomes, where is the transmitted vector), and uses a Gaussian approximation on the off-diagonal terms of the matrix. We also propose a simple estimation scheme which directly obtains an estimate of (instead of an estimate of), which is used as an effective channel estimate in the MPD algorithm. We refer to this receiver as the channel hardening-exploiting message passing (CHEMP) receiver. The proposed CHEMP receiver achieves very good performance in large-scaleMIMO systems (e.g., in systems with 16 to 128 uplink users and 128 base station antennas). For the considered large MIMO settings, the complexity of the proposed MPD algorithm is almost the same as or less than that of the minimum mean square error (MMSE) detection. This is because the MPD algorithm does not need a matrix inversion. It also achieves a significantly better performance compared to MMSE and other message passing detection algorithms using MMSE estimate of. Further, we design optimized irregular low density parity check (LDPC) codes specific to the considered large MIMO channel and the CHEMP receiver through EXIT chart matching. The LDPC codes thus obtained achieve improved coded bit error rate performance compared to off-the-shelf irregular LDPC codes.

Fault Tolerance on Large Scale Systems using Adaptive Process Replication

Exascale systems of the future are predicted to have mean time between failures (MTBF) of less than one hour. At such low MTBFs, employing periodic checkpointing alone will result in low efficiency because of the high number of application failures resulting in large amount of lost work due to rollbacks. In such scenarios, it is highly necessary to have proactive fault tolerance mechanisms that can help avoid significant number of failures. In this work, we have developed a mechanism for proactive fault tolerance using partial replication of a set of application processes. Our fault tolerance framework adaptively changes the set of replicated processes periodically based on failure predictions to avoid failures. We have developed an MPI prototype implementation, PAREP-MPI that allows changing the replica set. We have shown that our strategy involving adaptive process replication significantly outperforms existing mechanisms providing up to 20 percent improvement in application efficiency even for exascale systems.

Generalized Spatial Modulation in Large-Scale Multiuser MIMO Systems

Generalized spatial modulation (GSM) uses n(t) transmit antenna elements but fewer transmit radio frequency (RF) chains, n(rf). Spatial modulation (SM) and spatial multiplexing are special cases of GSM with n(rf) = 1 and n(rf) = n(t), respectively. In GSM, in addition to conveying information bits through n(rf) conventional modulation symbols (for example, QAM), the indices of the n(rf) active transmit antennas also convey information bits. In this paper, we investigate GSM for large-scale multiuser MIMO communications on the uplink. Our contributions in this paper include: 1) an average bit error probability (ABEP) analysis for maximum-likelihood detection in multiuser GSM-MIMO on the uplink, where we derive an upper bound on the ABEP, and 2) low-complexity algorithms for GSM-MIMO signal detection and channel estimation at the base station receiver based on message passing. The analytical upper bounds on the ABEP are found to be tight at moderate to high signal-to-noise ratios (SNR). The proposed receiver algorithms are found to scale very well in complexity while achieving near-optimal performance in large dimensions. Simulation results show that, for the same spectral efficiency, multiuser GSM-MIMO can outperform multiuser SM-MIMO as well as conventional multiuser MIMO, by about 2 to 9 dB at a bit error rate of 10(-3). Such SNR gains in GSM-MIMO compared to SM-MIMO and conventional MIMO can be attributed to the fact that, because of a larger number of spatial index bits, GSM-MIMO can use a lower-order QAM alphabet which is more power efficient.