A data restoration method for the large-scale disasters is proposed. Recently, social data, e.g. resident information or medical information, are stored into information systems. The social data are backed up to remote site to protect from data loss. It is important for restoration to be able to utilize the backup data as soon as possible. Thus, a part of backup data, e.g. files, is individually restored (File-lvl Restore). To reduce the time to get backup data by the File-lvl Restore more further, we apply the parallel data downloading. However, the current method does not consider the TCP slow-start processing; thus, the number of chunks of parallel downloading becomes insufficient. In this paper, we propose a data restoration method that decides the number of chunks based on Round Trip Time and the characteristics of TCP slow-start. Our evaluation results show that the time to get the backup data by using the proposed method becomes 1/2 to 1/3 compared to the current method.
In this paper, the parallel data restoration is evaluated with assumption of the storage node restart trend under large-scale disaster. In our previous study, we evaluated a proposed method for the data replication and data restoration for large-scale disasters. The proposed data replication copies data among nearby storage nodes each other to ensure reachability. Additionally, to broaden throughput, the proposed data restoration gets the backup data in parallel. The erasure coding is applied to avoid data lost. In this way, the data of the broken storage nodes are restored as soon as possible. However, the evaluations of previous study do not cover the transitional situation of recovery. After large-scale disaster, it is supposed that a part of storage nodes destructs and some storage nodes widely stop due to blackout. After a while, the storage nodes that temporally stop restart gradually. Thus, the number of the working storage nodes gradually increases. Therefore, a challenge of the parallel data restoration is to examine the relation between the suppression of the load and the prolongation of the restoration execution time under the situation and reveal the appropriate configurations. In this study, the model and method for the evaluation of the parallel data restoration are created. Additionally, the load and the restoration execution time are evaluated. From the results, the configuration of two data chunks and 30 parity chunks is suitable when the throughput is high. Meanwhile, the configuration of eight data chunks and 24 parity chunks is suitable when the throughput is low and the gradient of the storage node restart trend is steep. Moreover, if the gradient of the storage node restart trend is gradual, then for each throughput, the PDR configuration with the smallest number of data chunks of the configurations that can restore backup data is suitable.
Micromagnetic model simulations were performed to investigate spin-torque oscillators (STOs) integrated into the main poletrailing shield (MP-TS) gap of the write head (integrated STO). First, the oscillation states of a tilted STO were investigated in a model comprising the write head, STO, and medium soft underlayer (SUL). We found that inserting the STO into a non-parallel MP-TS gap resulted in more stable oscillation. Second, we performed full micromagnetic simulations of the write head, STO, SUL, and recording layer (RL), and the oscillation of the tilted STO was investigated. We found that the RL affected the STO oscillation to some extent; however, the effect was smaller than that from the write head. We also discuss the calculation times for various structures with and without the RL, and with/without medium motion.
Micromagnetic simulations were carried out to obtain stable oscillation of a spin transfer torque oscillator (STO) inserted into the gap between the main pole and trailing shield of a write head. We assumed a spin injection layer with in-plane anisotropy. Results show that reducing the magnetostatic interaction between the write head and the STO is the key to obtaining stable STO oscillation.
Microwave-assisted magnetic recording was investigated using a planar write head and exchange-coupled composite (ECC) media. When recording on ECC media using a planar head field distribution and the high frequency field generated by a spin torque oscillator it was possible to switch the media magnetisation into the opposite direction to the head field, i.e. the media effectively had a negative coercive field. The conditions for this effect to occur are discussed.
Antiferromagnetically coupled (AFC) media were modeled as a part of a microwave-assisted magnetic recording system. Recording media grains with magnetically hard and soft layers were AFC by a Ru interlayer. The magnetostatic field arising from such grains has a smaller influence on neighboring grains and recording layers during recording. Compared with single-layer recording media, the SNR of AFC media decreased much more slowly with increasing linear density, and AFC media SNR was superior at high densities. An example of an AFC medium with two recording layers is also presented.
With the increasing need for effective storage management due to ever-growing contentgeneration over the Internet, Distributed Storage Systems (DSS) has arisen as a valuable tool.Although DSS has considerably improved in the past years, it still leverages legacy techniques in its networking.To cope with the demanding requirements, Software Defined Networking (SDN) has revolutionized the way we manage networks and can significantly help in improving DSS network management.In this paper, we propose an SDN-based network control method that is capable of handling DSS network management and improving its performance.This paper presents the design, implementation, and evaluation using an emulated environment of a typical Data Center Network (DCN) deployment.The experiment results show that by applying the proposed method, DSS can increase the performance and service resilience compared to existing solutions.
Stable oscillation is one of the most important factors for spin-torque oscillators (STOs) used in a microwave-assisted magnetic recording system. We use a micromagnetic, integrated STO model, in which the STO is inserted into the main pole-trailing shield (MP-TS) gap of a write head, and show that the rise time of the in-gap field acting on the STO is critical to stable STO oscillation. We also show that the combination of a tilted STO and a tilted MP-TS gap results in stable STO oscillation due to weaker magnetostatic interactions between the STO and the write head.
Due to the increasing need for timely and reliable access to user-generated content, Distributed Storage Systems (DSS) became more relevant in the past years. However, since they have more interconnections among layers compared to traditional network applications, load imbalance issues arise. In this paper, we propose a hybrid approach combining server and link load balancing for multipath routing in DSS. The approach is Software Defined Networking (SDN) based, and uses a process we call on-demand inverse multiplexing. Preliminary results show that, by applying the proposal, the overall throughput considerably increases and resource usage remain balanced.
Selective recording on media with two discrete recording layers is complicated by magnetostatic interactions between the layers. Introducing antiferromagnetic exchange coupling can compensate for these interactions to some extent. This paper investigates the effect of the antiferromagnetic exchange coupling strength in a dual-layer medium and shows that the recording performance can be improved when the antiferromagnetic exchange coupling is optimized.
In recent years, shingled magnetic recording (SMR) has attracted attention as a new recording system of the hard disk drives. Although SMR can realize narrow track by overwriting data track like tile roofing, it has remarkable performance deterioration due to the influence of inter-track interference. Therefore, the performance improvement of the low-density parity-check (LDPC) coding and iterative decoding system using the sum-product decoding is required. In this paper, we propose the LDPC encoding and iterative decoding system with the log-likelihood ratio modulator and show the error floor is eliminated and improvement of signal-to-noise ratio to achieve the error free is obtained in the SMR under an areal recording density of 4 Tbits/in 2 specification for the channel bit sequence coded by run-length limited and LDPC encoders by computer simulation. Therefore, the channel bit is composed bit about five magnetic grains.
Simulations of recording on media with two and three distinct recording layers are presented. A high frequency field is used to select the layer on which data are recorded. Layer-selective recording was confirmed for two layer granular media with narrow distributions. For three layer media magnetostatic interactions between the layers led to a loss of selectivity in the bottom recording layer. Readout of data is possible by staggering transitions in different layers or by using a spin-torque oscillator.
The use of Distributed Storage Systems (DSS) has considerably increased in the past years, alongside the need for effective data transfer from storage to storage. Although current network infrastructure can reliably handle large amounts of traffic, networking techniques have not changed for several years, leading to an under-use of resources, i. e. most routing solutions still use single-path routing. In this paper, we present a pragmatic approach for multipath routing in DSS, which is based on Software Defined Networking (SDN) that uses parallel links at the edge-side. Path discovery is calculated by finding the k-maximum disjoint paths in a multigraph. Preliminary results show that, by using our multipath solution, not only the overall throughput increases but also the efficiency of resources usage.
The shingled magnetic recording (SMR) system is influenced by not only jitter-like medium noise from adjacent bits in the down-track direction but also intertrack interference (ITI) from the adjacent tracks. In this paper, we investigate the relationship between the recording pattern and the decoding reliability by classifying the log-likelihood ratio at the a posteriori probability decoder output according to the recording patterns under an SMR R-W channel specification of 4 Tb/in(2) by computer simulation. The result shows that the decoding reliability for the recording pattern "010" heavily affected by transition jitter is the lowest and little improved even if the influence of ITI is mitigated by a 2-D finite-impulse response filter.
Two-dimensional magnetic recording (TDMR), which combines shingled magnetic recording (SMR) with 2-D signal processing attracts attention from the hard disk drive industry. SMR is a magnetic recording system in order to increase the track density by writing tracks without the guard band. Therefore, the influences of writing and reading intertrack interferences (ITIs) deteriorate the quality of reproducing waveforms. Additionally, degradation of track due to overwriting and crosstalk from the adjacent tracks due to the width of reader sensitivity. The equalization using 2-D finite impulse response filter for the waveforms from three adjacent tracks is proposed as a method to reduce the influence of reading ITIs. In this paper, we evaluate the bit aspect ratio (BAR) in TDMR read/write channel under a specification of 4 Tb/in 2 using bit error rate of the low-density parity-check (LDPC) coding and iterative decoding system obtained by computer simulation. The results show that the suitable BAR in array head reading is 2.0.
A micromagnetic model analysis was carried out considering two types of write heads and three types of spin-torque oscillators (STOs). For write heads, we considered a small-sized head to reduce calculation time and a large-sized head with dimensions close to commercial heads. For STOs, we used a double-layered STO with reflection spin torque, a double-layered STO with transmission spin torque, and a tri-layered STO with two spin injection layers. We also carried out micromagnetic simulations of the STOs without the write heads (isolated STO) to compare the oscillation states. It was found that, compared with isolated STOs in a uniform applied field, a larger applied field was necessary for STOs integrated into a write head. This result shows that the write head should be designed to generate a large, well-shaped gap field to ensure stable STO oscillation.
Two-dimensional magnetic recording (TDMR) by shingled magnetic recording (SMR) draws attention as a next generation technology to increase the recording density in hard disk drive (HDD). It is shown that the two-dimensional finite impulse response (2D-FIR) filter provides gain the reproducing waveforms from an array head with 3 readers in the TDMR under a specification of 4 Tbit/inch.2 We evaluate the effect of the intertrack interference (ITI) reduction by 2D-FIR using the correlation between the real FIR filter output and the partial response class-I (PR1) signal corresponding to recording sequence by the computer simulation. The results show that the 2D-FIR filter is effective to mitigate ITI and improves the signal-to-noise ratio at the discriminate point by about 0.6 dB.