The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Les temps de réponse des disques SSD (Solid State Drives) ont considérablement diminué en raison de l'utilisation croissante de dispositifs à mémoire non volatile express (NVMe). De tels appareils ont des temps de réponse inférieurs à 100 microsecondes en moyenne. Les temps de réponse des systèmes 20 % Flash ont également considérablement diminué grâce à l’utilisation de disques SSD NVMe. Cependant, certaines applications, notamment les infrastructures de bureaux virtuels et les systèmes de bases de données en mémoire, nécessitent des systèmes de stockage avec des temps de réponse encore plus courts. Leurs charges de travail ont tendance à contenir de nombreuses concentrations d’entrées-sorties (IO), qui sont des agrégations d’accès IO. Ils ciblent des régions étroites du volume de stockage et peuvent durer jusqu’à une heure. Ces régions étroites occupent quelques pour cent de la capacité du nombre d'unités logiques, sont la cible de la plupart des accès IO et apparaissent à des adresses de blocs logiques imprévisibles. Pour réduire considérablement les temps de réponse de ces charges de travail, nous avons développé un système de stockage hiérarchisé automatisé appelé « stockage hiérarchisé automatisé avec mémoire rapide et stockage flash lent » (ATSMF) dans lequel les données des régions ciblées sont migrées entre les périphériques de stockage en fonction de la quantité restante prévue. durée de la concentration. L'environnement supposé est un serveur avec une mémoire non volatile et des disques SSD directement connectés, avec les applications utilisateur exécutées sur le serveur car cela réduit le temps de réponse moyen. Notre système prédit l'effet de la migration en utilisant les valeurs précédemment surveillées de l'augmentation du temps de réponse pendant la migration et de la modification du temps de réponse après la migration. Ces valeurs sont cohérentes pour chaque type de charge de travail si le système est construit à l'aide de mémoire non volatile et de disques SSD. En particulier, le système prédit la durée restante d'une concentration d'E/S, calcule l'augmentation attendue du temps de réponse pendant la migration et la diminution attendue du temps de réponse après la migration, et migre les données dans les régions ciblées si la somme des diminutions du temps de réponse après la migration. la migration dépasse la somme de l’augmentation du temps de réponse pendant la migration. Les résultats expérimentaux indiquent que l'ATSMF est au moins 50 % plus rapide que le stockage flash uniquement et que son taux d'accès à la mémoire est supérieur à XNUMX %.
Kazuichi OE
FUJITSU LABORATORIES LTD.
Mitsuru SATO
FUJITSU LABORATORIES LTD.
Takeshi NANRI
Kyushu University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Kazuichi OE, Mitsuru SATO, Takeshi NANRI, "ATSMF: Automated Tiered Storage with Fast Memory and Slow Flash Storage to Improve Response Time with Concentrated Input-Output (IO) Workloads" in IEICE TRANSACTIONS on Information,
vol. E101-D, no. 12, pp. 2889-2901, December 2018, doi: 10.1587/transinf.2018PAP0005.
Abstract: The response times of solid state drives (SSDs) have decreased dramatically due to the growing use of non-volatile memory express (NVMe) devices. Such devices have response times of less than 100 micro seconds on average. The response times of all-flash-array systems have also decreased dramatically through the use of NVMe SSDs. However, there are applications, particularly virtual desktop infrastructure and in-memory database systems, that require storage systems with even shorter response times. Their workloads tend to contain many input-output (IO) concentrations, which are aggregations of IO accesses. They target narrow regions of the storage volume and can continue for up to an hour. These narrow regions occupy a few percent of the logical unit number capacity, are the target of most IO accesses, and appear at unpredictable logical block addresses. To drastically reduce the response times for such workloads, we developed an automated tiered storage system called “automated tiered storage with fast memory and slow flash storage” (ATSMF) in which the data in targeted regions are migrated between storage devices depending on the predicted remaining duration of the concentration. The assumed environment is a server with non-volatile memory and directly attached SSDs, with the user applications executed on the server as this reduces the average response time. Our system predicts the effect of migration by using the previously monitored values of the increase in response time during migration and the change in response time after migration. These values are consistent for each type of workload if the system is built using both non-volatile memory and SSDs. In particular, the system predicts the remaining duration of an IO concentration, calculates the expected response-time increase during migration and the expected response-time decrease after migration, and migrates the data in the targeted regions if the sum of response-time decrease after migration exceeds the sum of response-time increase during migration. Experimental results indicate that ATSMF is at least 20% faster than flash storage only and that its memory access ratio is more than 50%.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018PAP0005/_p
Copier
@ARTICLE{e101-d_12_2889,
author={Kazuichi OE, Mitsuru SATO, Takeshi NANRI, },
journal={IEICE TRANSACTIONS on Information},
title={ATSMF: Automated Tiered Storage with Fast Memory and Slow Flash Storage to Improve Response Time with Concentrated Input-Output (IO) Workloads},
year={2018},
volume={E101-D},
number={12},
pages={2889-2901},
abstract={The response times of solid state drives (SSDs) have decreased dramatically due to the growing use of non-volatile memory express (NVMe) devices. Such devices have response times of less than 100 micro seconds on average. The response times of all-flash-array systems have also decreased dramatically through the use of NVMe SSDs. However, there are applications, particularly virtual desktop infrastructure and in-memory database systems, that require storage systems with even shorter response times. Their workloads tend to contain many input-output (IO) concentrations, which are aggregations of IO accesses. They target narrow regions of the storage volume and can continue for up to an hour. These narrow regions occupy a few percent of the logical unit number capacity, are the target of most IO accesses, and appear at unpredictable logical block addresses. To drastically reduce the response times for such workloads, we developed an automated tiered storage system called “automated tiered storage with fast memory and slow flash storage” (ATSMF) in which the data in targeted regions are migrated between storage devices depending on the predicted remaining duration of the concentration. The assumed environment is a server with non-volatile memory and directly attached SSDs, with the user applications executed on the server as this reduces the average response time. Our system predicts the effect of migration by using the previously monitored values of the increase in response time during migration and the change in response time after migration. These values are consistent for each type of workload if the system is built using both non-volatile memory and SSDs. In particular, the system predicts the remaining duration of an IO concentration, calculates the expected response-time increase during migration and the expected response-time decrease after migration, and migrates the data in the targeted regions if the sum of response-time decrease after migration exceeds the sum of response-time increase during migration. Experimental results indicate that ATSMF is at least 20% faster than flash storage only and that its memory access ratio is more than 50%.},
keywords={},
doi={10.1587/transinf.2018PAP0005},
ISSN={1745-1361},
month={December},}
Copier
TY - JOUR
TI - ATSMF: Automated Tiered Storage with Fast Memory and Slow Flash Storage to Improve Response Time with Concentrated Input-Output (IO) Workloads
T2 - IEICE TRANSACTIONS on Information
SP - 2889
EP - 2901
AU - Kazuichi OE
AU - Mitsuru SATO
AU - Takeshi NANRI
PY - 2018
DO - 10.1587/transinf.2018PAP0005
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E101-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2018
AB - The response times of solid state drives (SSDs) have decreased dramatically due to the growing use of non-volatile memory express (NVMe) devices. Such devices have response times of less than 100 micro seconds on average. The response times of all-flash-array systems have also decreased dramatically through the use of NVMe SSDs. However, there are applications, particularly virtual desktop infrastructure and in-memory database systems, that require storage systems with even shorter response times. Their workloads tend to contain many input-output (IO) concentrations, which are aggregations of IO accesses. They target narrow regions of the storage volume and can continue for up to an hour. These narrow regions occupy a few percent of the logical unit number capacity, are the target of most IO accesses, and appear at unpredictable logical block addresses. To drastically reduce the response times for such workloads, we developed an automated tiered storage system called “automated tiered storage with fast memory and slow flash storage” (ATSMF) in which the data in targeted regions are migrated between storage devices depending on the predicted remaining duration of the concentration. The assumed environment is a server with non-volatile memory and directly attached SSDs, with the user applications executed on the server as this reduces the average response time. Our system predicts the effect of migration by using the previously monitored values of the increase in response time during migration and the change in response time after migration. These values are consistent for each type of workload if the system is built using both non-volatile memory and SSDs. In particular, the system predicts the remaining duration of an IO concentration, calculates the expected response-time increase during migration and the expected response-time decrease after migration, and migrates the data in the targeted regions if the sum of response-time decrease after migration exceeds the sum of response-time increase during migration. Experimental results indicate that ATSMF is at least 20% faster than flash storage only and that its memory access ratio is more than 50%.
ER -