Search

The demand for high performance I/O in Storage-as-a-Service (SaaS) is increasing day by day. To address this demand, NAND Flash-based Solid-state Drives (SSDs) are commonly used in data centers as cache- or top-tiers in the storage rack ascribe to their superior performance compared to traditional hard disk drives (HDDs). Meanwhile, with the capital expenditure of SSDs declining and the storage capacity of SSDs increasing, all-flash data centers are evolving to serve cloud services better than SSD-HDD hybrid data centers. During this transition, the biggest challenge is how to reduce the Write Amplification Factor (WAF) as well as to improve the endurance of SSD since this device has a limited program/erase cycles. A specified case is that storing data with different lifetimes (i.e., I/O streams with similar temporal fetching patterns such as reaccess frequency) in one single SSD can cause high WAF, reduce the endurance, and downgrade the performance of SSDs. Motivated by this, multi-stream SSDs have been developed to enable data with a different lifetime to be stored in different SSD regions. The logic behind this is to reduce the internal movement of data—when garbage collection is triggered, there are high chances of having data blocks with either all the pages being invalid or valid. However, the limitation of this technology is that the system needs to manually assign the same streamID to data with a similar lifetime. Unfortunately, when data arrives, it is not known how important this data is and how long this data will stay unmodified. Moreover, according to our observation, with different definitions of a lifetime (i.e., different calculation formulas based on selected features previously exhibited by data, such as sequentiality, and frequency), streamID identification may have varying impacts on the final WAF of multi-stream SSDs. Thus, in this article, we first develop a portable and adaptable framework to study the impacts of different workload features and their combinations on write amplification. We then propose a feature-based stream identification approach, which automatically co-relates the measurable workload attributes (such as I/O size, I/O rate, and so on.) with high-level workload features (such as frequency, sequentiality, and so on.) and determines a right combination of workload features for assigning streamIDs . Finally, we develop an adaptable stream assignment technique to assign streamID for changing workloads dynamically. Our evaluation results show that our automation approach of stream detection and separation can effectively reduce the WAF by using appropriate features for stream assignment with minimal implementation overhead.

In the era of big data processing, more and more data centers in cloud storage are now replacing traditional HDDs with enterprise SSDs. Both developers and users of these SSDs require thorough benchmarking to evaluate and configure the variable parameters of emerging technologies. Multi-stream SSD [2], [3] is the recent development of the SSD industry, which assists in placing data on SSDs in a smart way to improve application performance and SSD endurance. The challenging part to use multi-stream SSDs is to assign stream IDs to incoming writes, such that each stream consists of data with a similar lifetime. The benefit of the stream management algorithms varies over different workloads. Thus, first, we propose a new framework, called Pattern I/O generator (PatIO), to capture the enterprise storage behavior that is prevailing across various user workloads, virtualization setup, file systems, and volume managers for the database server applications on flash-based storage. Second, using PatIO, we study what type of applications may be benefited by which stream assignment algorithm. Third, we design the framework to automatically tune the variable parameters of different stream identification algorithms of the multi-stream SSDs. Our evaluation shows 20% to 110% of the reward function increase, measuring the cumulative impact on application performance and SSD endurance.

An increasing concern that curbs the widespread adoption of KV-SSD is whether or not offloading host-side operations to the storage device changes device behavior, negatively affecting various applications’ overall performance. In this paper, we systematically measure, quantify, and understand the performance of KV-SSD by studying the impact of its distinct components such as indexing, data packing, and key handling on I/O concurrency, garbage collection, and space utilization. Our experiments and analysis uncover that KV-SSD’s behavior differs from well-known idiosyncrasies of block-SSD. Proper understanding of its characteristics will enable us to achieve better performance for random, read-heavy, and highly concurrent workloads.

Computer games are an extremely popular but overlooked workload. Cloud-gaming has been one of the biggest buzzwords in the gaming industry throughout 2020. The rapid growth of the video gaming industry and the diverse set of popular video games available today raises increasing concern to properly understand its I/O characteristics to improve their performance and design better gaming servers and consoles. To the best of our knowledge, this is the first attempt to systematically measure, quantify, and characterize the organization of game data into files, back-end storage access patterns, and the performance of gaming workloads. We explore the I/O behavior of 14 recent and famous games, producing a series of observations coming from measurements done on a real setup.

The development of KV-SSDs allows simplifying the I/O stack compared to the traditional block-based SSDs. We propose a novel Key-Value-based Storage infrastructure for Parallel Computing(KV-SiPC)-a framework for multi-thread OpenMP applications to use NVMe-based KV-SSDs. We design a new capability to execute workloads with multiple parallel data threads along with traditional parallel compute threads, that allow us to improve the overall throughput of applications, utilizing the maximum possible storage bandwidth. We implement our KV-SiPC infrastructure in a real system by extending various processing layers (e.g., program, OS, and device layers) and evaluate the performance of KV-SiPC by using block-based NVMe SSDs in the traditional I/O stack as a baseline for comparisons. The experimental results show that KV-SiPC can better utilize the available device bandwidth and significantly increases application I/O throughput.

Predicting the performance of an application running on parallel computing platforms is increasingly becoming important because of its influence on development time and resource management. However, predicting the performance with respect to parallel processes is complex for iterative and multi-stage applications. This research proposes a performance approximation approach FiM to predict the calculation time with FiM-Cal and communication time with FiM-Com of an application running on a distributed framework. FiM-Cal consists of two key components that are coupled with each other: (1) a Stochastic Markov Model to capture non-deterministic runtime that often depends on parallel resources, e.g., number of processes, and (2) a machine-learning model that extrapolates the parameters for calibrating our Markov model when we have changes in application parameters such as dataset. Along with the parallel calculation time, parallel computing platforms consume some data transfer time to communicate among different nodes. FiM-Com consists of a simulation queuing model to quickly estimate communication time. Our new modeling approach considers different design choices along multiple dimensions, namely (i) process-level parallelism, (ii) distribution of cores on multi-processor platform, (iii) application related parameters, and (iv) characteristics of datasets. The major contribution of our prediction approach is that FiM can provide an accurate prediction of parallel processing time for the datasets that have a much larger size than that of the training datasets. We evaluate our approach with NAS Parallel Benchmarks and real iterative data processing applications. We compare the predicted results (e.g., end-to-end execution time) with actual experimental measurements on a real distributed platform. We also compare our work with an existing prediction technique based on machine learning. We rank the number of processes according to the actual and predicted results from FiM and calculate the correlation between the actual and predicted rankings. Our results show that FiM obtains a high correlation in the range of 0.80 to 0.99, which indicates considerable accuracy of our technique. Such prediction provides data analysts a useful insight of optimal configuration of parallel resources (e.g., number of processes and number of cores) and also helps system designers to investigate the impact of changes in application parameters on system performance.

By using fast back-end storage, performance benefits of a lightweight container platform can be leveraged with quick I/O response. Nevertheless, the performance of simultaneously executing multiple instances of same or different applications may vary significantly with the number of containers. The performance may also vary with the nature of applications because different applications can exhibit different nature on SSDs in terms of I/O types (read/write), I/O access pattern (random/sequential), I/O size, etc. Therefore, this paper aims to investigate and analyze the performance characterization of both homogeneous and heterogeneous mixtures of I/O intensive containerized applications, operating with high performance NVMe SSDs and derive novel design guidelines for achieving an optimal and fair operation of the both homogeneous and heterogeneous mixtures. By leveraging these design guidelines, we further develop a new docker controller for scheduling workload containers of different types of applications. Our controller decides the optimal batches of simultaneously operating containers in order to minimize total execution time and maximize resource utilization. Meanwhile, our controller also strives to balance the throughput among all simultaneously running applications. We develop this new docker controller by solving an optimization problem using five different optimization solvers. We conduct our experiments in a platform of multiple docker containers operating on an array of three enterprise NVMe drives. We further evaluate our controller using different applications of diverse I/O behaviors and compare it with simultaneous operation of containers without the controller. Our evaluation results show that our new docker workload controller helps speed-up the overall execution of multiple applications on SSDs., Janki Bhimani,Zhengyu Yang,Ningfang Mi,Jingpei Yang,Jingpei Yang,Manu Awasthi, Rajinikanth Pandurangan,Vijay Balakrishnan

In the era of big data processing, more and more datacenters in cloud storages are now replacing traditional HDDs with enterprise SSDs. Both developers and users of these SSDs require thorough benchmarking to evaluate their performance impacts. I/O performance with synthetic workload or classic benchmark varies drastically from real I/O activities in the datacenter. Thus, we propose a new framework, called Pattern I/O generator (PatIO), to collectively capture the enterprise storage behavior that is prevailing across assorted user workloads and system configurations for different database server applications on flash-based storage. PatIO is designed to emulate the processing of real-world I/O activities easily with less time and resource requirements. Our methodology comprises three main steps: (1) dissect the overall I/O activities of various real workloads and identify the prevailing attributes in distinct visual I/O patterns; (2) construct a pattern warehouse as the collection of unique I/O patterns that are generated through various combinations of multiple I/O jobs; and (3) finally integrate different combinations of these synthetically generated I/O patterns to reproduce the comprehensive characteristics of various real workloads and system setup for the database server applications. To provide an easy-to-use experience, we develop a graphical user interface (GUI). We evaluate our framework by comparing I/O characteristics and I/O performance of generated workloads with those of real-world workloads for multiple database applications such as MySQL, Cassandra, and ForestDB.

Apache Spark is an in-memory analytic framework that has been adopted in the industry and research fields. Two memory managers, Static and Unified, are available in Spark to allocate memory for caching Resilient Distributed Datasets (RDDs) and executing tasks. However, we found that the static memory manager (SMM) lacks flexibility, while the unified memory manager (UMM) puts heavy pressure on the garbage collection of JVM on which Spark resides. To address these issues, we design an auto-tuning memory manager (ATuMm) to support dynamic memory allocation with the consideration of both memory demands and latency introduced by garbage collection. We implement our new memory manager in Spark 2.2.0 and evaluate it by conducting experiments in a real Spark cluster. Our experimental results show that our auto-tuning memory manager can reduce the total garbage collection time and thus further improve the performance (i.e., reduced latency) of Spark applications, compared to the existing Spark memory management solutions.

Vibration generated in modern computing environments such as autonomous vehicles, edge computing infrastructure, and data center systems is an increasing concern. In this paper, we systematically measure, quantify and characterize the impact of vibration on the performance of SSD devices. Our experiments and analysis uncover that exposure to both short-term and long-term vibration, even within the vendor-specified limits, can significantly affect SSD I/O performance and reliability.

The demand for high speed 'Storage-as-a-Service' (SaaS) is increasing day-by-day. SSDs are commonly used in higher tiers of storage rack in data centers. Also, all flash data centers are evolving to better serve cloud services. Although SSDs guaranty better performance when compared to HDDs, but SSDs endurance is still a matter of concern. Storing data with different lifetime in an SSD can cause high write amplification and reduce the endurance and performance of SSDs. Recently, multi-stream SSDs have been developed to enable data with different lifetime to be stored in different SSD regions and thus reduce write amplification. To efficiently use this new multi-streaming technology, it is important to choose appropriate workload features to assign the same streamID to data with similar lifetime. However, we found that streamID identification using different features may have varying impacts on the final write amplification of multi-stream SSDs. Therefore, in this paper we develop a portable and adoptable framework to study the impacts of different workload features and their combinations on write amplification. We also introduce a new feature, named "coherency", to capture the friendship among write operations with respect to their update time. Finally, we propose a feature-based stream identification approach, which co-relates the measurable workload attributes (such as I/O size, I/O rate, etc.) with high level workload features (such as frequency, sequentiality etc.) and determines a good combination of workload features for assigning streamIDs. Our evaluation results show that our proposed approach can always reduce the Write Amplification Factor (WAF) by using appropriate features for stream assignment.

Data temperature identification is an importance issue of many fields like data caching and storage tiering in modern flash-based storage systems. With the technological advancement of memory and storage, data temperature identification is no longer just a classification of hot and cold, but instead becomes a "multistreaming" data categorization problem to classify data into multiple categories according to their temperature. Therefore, we propose a novel data temperature identification scheme that adopts bloom filters to efficiently capture both frequency and recency of data blocks and accurately identify the exact data temperature for each data block. Moreover, in bloom filter data structure we replace the original OR operation with the XOR masking operation such that our scheme can delete or reset bits in bloom filters and thus avoid high false positives due to saturation. We further utilize twin bloom filters to alternatively keep unmasked clean copies of data and thus ensure low false negative rate. Our extensive evaluation results show that our new scheme can accurately identify the exact data temperature with low false identification rates across different synthetic and real I/O workloads. More importantly, our scheme consumes less memory space compared to other existing data temperature identification schemes.

Hadoop YARN is an Apache Software Foundation open project that provides a resource management framework for large scale parallel data processing, such as MapReduce jobs. Fair scheduler is a dispatcher which has been widely used in YARN to assign resources fairly and equally to applications. However, there exists a problem of the Fair scheduler when the resource requisition of applications is beyond the amount that the cluster can provide. In such a case, the YARN system will be halted if all resources are occupied by ApplicationMasters, a special task of each job that negotiates resources for processing tasks and coordinates job execution. To solve this problem, we propose an automatic and dynamic admission control mechanism to prevent the ceasing situation happened when the requested amount of resources exceeds the cluster resource capacity, and dynamically reserve resources for processing tasks in order to obtain good performance, e.g., reducing makespans of MapReduce jobs. After collecting resource usage information of each work node, our mechanism dynamically predicts the amount of reserved resources for processing tasks and automatically controls running jobs based on the prediction. We implement the new mechanism in Hadoop YARN and evaluate it with representative MapReduce benchmarks. The experimental results show the effectiveness and robustness of this mechanism under both homogeneous and heterogeneous workloads.

Recently, the capital expenditure of flash-based Solid State Driver (SSDs) keeps declining and the storage capacity of SSDs keeps increasing. As a result, all-flash storage systems have started to become more economically viable for large shared storage installations in datacenters, where metrics like Total Cost of Ownership (TCO) are of paramount importance. On the other hand, flash devices suffer from write amplification, which, if unaccounted, can substantially increase the TCO of a storage system. In this paper, we first develop a TCO model for datacenter all-flash storage systems, and then plug a Write Amplification model (WAF) of NVMe SSDs we build based on empirical data into this TCO model. Our new WAF model accounts for workload characteristics like write rate and percentage of sequential writes. Furthermore, using both the TCO and WAF models as the optimization criterion, we design new flash resource management schemes (MINTCO) to guide datacenter managers to make workload allocation decisions under the consideration of TCO for SSDs. Based on that, we also develop MINTCO-RAID to support RAID SSDs and MINTCO-OFFLINE to optimize the offline workload-disk deployment problem during the initialization phase. Experimental results show that MINTCO can reduce the TCO and keep relatively high throughput and space utilization of the entire datacenter storage resources.

We propose a nearby traffic flow modelling solution based on built-in Cyber-Physical System (CPS) sensors of autonomous vehicles. Our goal is to enhance the offline route planning and driving decision adjustment based on the first-hand traffic information, especially during poor Internet connection moments. Specifically, our model helps to select the optimal speed on a road, the optimal distance for timing to brake, and the safe distance from other vehicles to keep. Moreover, our model can also assist neighboring autonomous vehicles by communicating required information through Ad-Hoc network communications or through a centralized cloud. In detail, we first focus on the unique characteristic of traffic flow (such as traffic rule, avoid collision behaviours), and then build a comprehensive model to handle multiple scenarios. Technically, our model uses density functions of velocities, the differential equation of traffic flows, and the traffic viscosity with information collected from the traffic flow, the distances between vehicles, the amount and density of vehicle, the instant velocity, the speed limit, and the momentum to analysis the the driving scene. We evaluate our model with real traffic data collected by in-vehicle CPS sensors to the proposed nearby traffic flow model. Results show that our work can accurately conduct offline estimation on nearby traffic signal influence, and reveal the correlations among velocity, density and (spatial and temporal) location to adjust route during runtime.

In the year of 2017, more and more datacenters have started to replace traditional SATA and SAS SSDs with NVMe SSDs due to NVMe's outstanding performance [1]. However, for historical reasons, current popular deployments of NVMe in VM-hypervisor-based platforms (such as VMware ESXi [2]) have numbers of intermediate queues along the I/O stack. As a result, performance is bottlenecked by synchronization locks in these queues, cross-VM interference induces I/O latency, and most importantly, up-to-64K-queue capability of NVMe SSDs cannot be fully utilized. In this paper, we developed a hybrid framework of NVMe-based storage system called “H-NVMe”, which provides two VM I/O stack deployment modes “Parallel Queue Mode” and “Direct Access Mode”. The first mode increases parallelism and enables lock-free operations by implementing local lightweight queues in the NVMe driver. The second mode further bypasses the entire I/O stack in the hypervisor layer and allows trusted user applications whose hosting VMDK (Virtual Machine Disk) files are attached with our customized vSphere IOFilters [3] to directly access NVMe SSDs to improve the performance isolation. This suits premium users who have higher priorities and the permission to attach IOFilter to their VMDKs. H-NVMe is implemented on VMware EXSi 6.0.0, and our evaluation results show that the proposed H-NVMe framework can significant improve throughputs and bandwidths compared to the original inbox NVMe solution.

In the year of 2017, the capital expenditure of Flash-based Solid State Drivers (SSDs) keeps declining and the storage capacity of SSDs keeps increasing. As a result, the “selling point” of traditional spinning Hard Disk Drives (HDDs) as a backend storage — low cost and large capacity — is no longer unique, and eventually they will be replaced by low-end SSDs which have large capacity but perform orders of magnitude better than HDDs. Thus, it is widely believed that all-flash multi-tier storage systems will be adopted in the enterprise datacenters in the near future. However, existing caching or tiering solutions for SSD-HDD hybrid storage systems are not suitable for all-flash storage systems. This is because that all-flash storage systems do not have a large speed difference (e.g., 10x) among each tier. Instead, different specialties (such as high performance, high capacity, etc.) of each tier should be taken into consideration. Motivated by this, we develop an automatic data placement manager called “AutoTiering” to handle virtual machine disk files (VMDK) allocation and migration in an all-flash multitier datacenter to best utilize the storage resource, optimize the performance, and reduce the migration overhead. AutoTiering is based on an optimization framework, whose core technique is to predict VM's performance change on different tiers with different specialties without conducting real migration. As far as we know, AutoTiering is the first optimization solution designed for all-flash multi-tier datacenters. We implement AutoTiering on VMware ESXi [1], and experimental results show that it can significantly improve the I/O performance compared to existing solutions.

Cyber-Physical System (CPS) is a rising technology that utilizes computation and storage resources for sensing, processing, analysis, predicting, understanding of field-data, and then uses communication resources for interaction, intervene, and interface management, and finally provides control for systems so that they can inter-operate, evolve, and run in a stable evidence-based environment. There are two major demands when building the storage infrastructure for a CPS cluster to support above-mentioned functionalities: (1) high I/O and network throughput requirements during runtime, and (2) low latency demand for disaster recovery. To address challenges brought by these demands, in this paper, we propose a complete solution called AutoReplica - an automatic and scalable data replication manager in distributed computation and storage infrastructure of cyber-physical systems, using tiering storage with SSD (solid state disk) and HDD (hard disk drive). Specifically, AutoReplica uses SSD to absorb hot data and to maximize I/Os, and its intelligent replication scheme further helps to recovery from disaster. To effectively balance the trade-off between I/O performance and fault tolerance, AutoReplica utilizes the SSDs of remote CPS server nodes (which are connected by high speed fibers) to replicate hot datasets cached in the SSD tier of the local CPS server node. AutoReplica has three approaches to build the replica cluster in order to support multiple SLAs. AutoReplica automatically balances loads among nodes, and can conduct seamlessly online migration operation (i.e., migrate-on-write scheme), instead of pausing the subsystem and copying the entire dataset from one node to the other. Lastly, AutoReplica supports parallel prefetching from both primary node and replica node(s) with a new dynamic optimizing streaming technique to improve I/O performance. We implemented AutoReplica on a real CPS infrastructure, and experimental results show that AutoReplica can significantly reduce the total recovery time with slight overhead compared to the no replication cluster and traditional replication clusters.