投稿時間:2022-10-21 07:21:07 RSSフィード2022-10-21 07:00 分まとめ(21件)

カテゴリー等	サイト名等	記事タイトル・トレンドワード等	リンクURL	頻出ワード・要約等/検索ボリューム	登録日
IT	InfoQ	Presentation: Securing Java Applications in the Age of Log4Shell	https://www.infoq.com/presentations/log4shell-security/?utm_campaign=infoq_content&utm_source=infoq&utm_medium=feed&utm_term=global	Presentation Securing Java Applications in the Age of LogShellSimon Maple looks at how one can be more proactive and defensive in decisions for future LogShell like scenarios considering identifying and reducing the risk introduced into applications By Simon Maple	2022-10-20 21:08:00
IT	ITmedia 総合記事一覧	[ITmedia News] 「Pixel Watch」をiFixitが分解　ケースを開けるのは簡単だが…	https://www.itmedia.co.jp/news/articles/2210/21/news073.html	google	2022-10-21 06:53:00
IT	ITmedia 総合記事一覧	[ITmedia ビジネスオンライン] お金が見る見る溶けていく――投資初心者がハマる「レバレッジ投信」、2022年は最悪のリターンに？	https://www.itmedia.co.jp/business/articles/2210/21/news045.html	itmedia	2022-10-21 06:30:00
IT	ビジネス+IT 最新ニュース	茨城最強カフェ「サザコーヒー」の独自戦略、世界のスタバに負けない3つのこだわり	https://www.sbbit.jp/article/cont1/95591?ref=rss	茨城最強カフェ「サザコーヒー」の独自戦略、世界のスタバに負けないつのこだわり人口約万人の茨城県ひたちなか市ー。	2022-10-21 06:10:00
AWS	AWS Partner Network (APN) Blog	Realizing Your Clean Energy Goals with Accenture’s Data-Led Transformation on AWS	https://aws.amazon.com/blogs/apn/realizing-your-clean-energy-goals-with-accenture-data-led-transformation-on-aws/	Realizing Your Clean Energy Goals with Accenture s Data Led Transformation on AWSWhile utilities have historically been rich with data from customers programs and assets many organizations often manage data in siloes Source data can also be disorganized with deficiencies in defined quality assurance and quality control processes Learn how utilities are successfully embracing Accenture s data led transformation DLT and leveraging accelerators powered by AWS to reach their business objectives and meet regulatory obligations	2022-10-20 21:32:19
AWS	AWS	Alexa, be quiet! End-to-end near-real time model building and evaluation in Amazon Alexa	https://www.youtube.com/watch?v=zP2SLReYmt0	Alexa be quiet End to end near real time model building and evaluation in Amazon AlexaTo improve Amazon Alexa experiences and support machine learning inference at scale we built an automated end to end solution for incremental model building or fine tuning machine learning models through continuous learning continual learning and or semi supervised active learning Customer privacy is our top concern at Alexa and as we build solutions we face unique challenges when operating at scale such as supporting multiple applications with tens of thousands of transactions per second with several dependencies including near real time inference endpoints at low latencies Apache Flink helps us transform and discover metrics in near real time in our solution In this talk we will cover the challenges that we faced how we scale the infrastructure to meet the needs of ML teams across Alexa and go into how we enable specific use cases that use Apache Flink on Amazon Kinesis Data Analytics to improve Alexa experiences to delight our customers while preserving their privacy Kinesis Data Analytics for Apache Flink Subscribe More AWS videos More AWS events videos ABOUT AWSAmazon Web Services AWS is the world s most comprehensive and broadly adopted cloud platform offering over fully featured services from data centers globally Millions of customers ーincluding the fastest growing startups largest enterprises and leading government agencies ーare using AWS to lower costs become more agile and innovate faster Flink AsyncSink KinesisDataAnalytics AWS AmazonWebServices CloudComputing	2022-10-20 21:20:22
AWS	AWSタグが付けられた新着投稿 - Qiita	MacにてEKSでGrafanaを利用する	https://qiita.com/miyabiz/items/ce782ac64592633dbf31	awscli	2022-10-21 06:51:29
海外TECH	DEV Community	Performance comparison: Reduct Storage vs. Minio	https://dev.to/reduct-storage/performance-comparison-reduct-storage-vs-minio-533e	Performance comparison Reduct Storage vs MinioWe often use blob storage like S If we need to store data of different formats and sizes somewhere in the cloud or in our internal storage Minio is a S compatible storage which you can run on your private cloud bare metal server or even on an edge device You can also adapt it to keep historical data as time series of blobs The most straightforward solution would be to create a folder for each data source and save objects with timestamps in their names bucket cv camera bin bin binIf you need to query data you should request a list of objects in the cv camera folder and filter them with names which are in the given time interval This approach is simple for implementation but it has some disadvantages the more objects the folder has the longer the querying is big overhead for small objects timestamps as strings and minimal file size is Kb or due to the block size of the file systemFIFO quota to remove old data when we reach some limit may not work for intensive write operations Reduct Storage aims to solve these issues It has a strong FIFO quota an HTTP API for querying data via time intervals and it composes objects or records into blocks for an efficient disk usage and search Minio and Reduct Storage have Python SDKs so we can use them for implementation write and read operations and compare the performance Read Write Data With MinioFor benchmarks we create two functions to write and read CHUNK COUNT chunks from minio import Minioimport timeminio client Minio access key minioadmin secret key minioadmin secure False def write to minio count for i in range CHUNK COUNT count CHUNK SIZE object name f data str int time time ns bin minio client put object BUCKET NAME object name io BytesIO CHUNK CHUNK SIZE return count count data to print it in main functiondef read from minio t t count t str int t t str int t for obj in minio client list objects test prefix data if t lt obj object name lt t resp minio client get object test obj object name count len resp read return countYou can that minio client doesn t provide any API query data with patterns so we have to browse the whole folder on the client side to find the needed object If you have billions of objects it stops working You should store object paths in some time series database or create a hierarchy of folders e g create a folder per day Read Write Data With Reduct StorageWith Reduct Storage this is a way easier from reduct import Client as ReductClientreduct client ReductClient async def write to reduct count bucket await reduct client create bucket test exist ok True for i in range CHUNK COUNT await bucket write data CHUNK count CHUNK SIZE return countasync def read from reduct t t count bucket await reduct client get bucket test async for rec in bucket query data int t int t count len await rec read all return count BenchmarksWhen we have the write read functions we can finally write our benchmarks import ioimport randomimport timeimport asynciofrom minio import Miniofrom reduct import Client as ReductClientCHUNK SIZE CHUNK COUNT BUCKET NAME test CHUNK random randbytes CHUNK SIZE minio client Minio access key minioadmin secret key minioadmin secure False reduct client ReductClient Our function were here if name main print f Chunk size CHUNK SIZE Mb count CHUNK COUNT ts time time size write to minio print f Write size Mb to Minio time time ts s ts read time time size read from minio ts time time print f Read size Mb from Minio time time ts read s loop asyncio new event loop ts time time size loop run until complete write to reduct print f Write size Mb to Reduct Storage time time ts s ts read time time size loop run until complete read from reduct ts time time print f Read size Mb from Reduct Storage time time ts read s For testings we need to run the databases It is easy to do with docker compose services reduct storage image reductstorage engine v volumes reduct data data ports minio image minio minio volumes minio data data command minio server data console address ports Run the docker compose configuration and the benchmarks docker compose up dpython main py ResultsThe script print the results for given CHUNK SIZE and CHUNK COUNT On my device I got the following numbers ChunkOperationMinioReduct Storage Mb requests Write s sRead s s Mb requests Write s sRead s s Mb requests Write s sRead s sAs you can see Reduct Storage is always faster for write operations times for Mb blobs and a bit slower for reading when we have many small objects You may notice that the speed decreasing for both databases when we reduce the size of chunks This can be explained with HTTP overhead because we spend a dedicated HTTP request for each write or read operation ConclusionsReduct Storage could be a good option for applications where you should store blobs historically with timestamps and write data all the time It has a strong FIFO quota to avoid problems with disk space and it is very fast for intensive write operations References Reduct StorageMinioReduct Storage Client SDK for PythonFull Example on GitHub	2022-10-20 21:56:45
Apple	AppleInsider - Frontpage News	Compared: New Apple TV 4K versus 2021 Apple TV 4K	https://appleinsider.com/articles/22/10/19/compared-new-apple-tv-4k-versus-2021-apple-tv-4k?utm_medium=rss	Compared New Apple TV K versus Apple TV KApple has updated the Apple TV K with a thinner design and an upgraded processor Here s how it compares to the older model Left Apple TV K Right Apple TV K The device is a modest update from the model that Apple released in The version is thinner and lighter and now comes in two configurations It also starts at a lower price point than the previous model Read more	2022-10-20 21:45:35
海外TECH	Engadget	GM officially reveals its $107,000 electrified 2024 GMC Sierra Denali	https://www.engadget.com/gm-officially-reveals-its-electrified-2024-gmc-sierra-denali-213601650.html?src=rss	GM officially reveals its electrified GMC Sierra DenaliGM s goal to sell nothing but EVs by is well on its way with massive demand already for the GMC Hummer EV and the Chevy Silverado The automaker took to social media on Thursday to unveil its third electrified offering and the burly looking Sierra Denali EV that the company revealed does not disappoint Introducing the First Ever GMCSierraEV Denali The Denali of EVs pic twitter com pXkeQIZbHーGMC GMC October Officially it s named the GMC Sierra EV Denali Edition and it ll start shipping in early with an MSRP of plus dealer fees The Sierra is built on the same Ultium battery technology as the Hummer and Silverado so GM estimates the Sierra will have plus miles of range on a full charge miles more than the Blazer EV announced in July offer HP lb ft torque inch rims and the ability to tow up to pounds nbsp On a DC fast charge the Sierra should be able to tack on miles of range with every ten minutes of station time ーassuming you spring for the W electrical architecture which is looking to be an optional feature along with four wheel steering and crab walking capabilities Like other Ultium vehicles the Sierra will offer bidirectional charging enabling it to power household appliances for up to days GM s product site reads A subscription for GM s SuperCruise hands free ADAS system will be included for three years as well The Denali is the first of three Sierra EV variants slated for release in the next few years The standard edition Sierra EV Elevation will arrive in early with inch rims The off road ready AT will arrive a year earlier in and offer two additional inches of ground clearance from the base Elevation	2022-10-20 21:36:01
海外ニュース	Japan Times latest articles	Liz Truss quits after six chaotic weeks as U.K. prime minister	https://www.japantimes.co.jp/news/2022/10/21/world/politics-diplomacy-world/truss-resign/	Liz Truss quits after six chaotic weeks as U K prime ministerLiz Truss resigns as prime minister after being brought down by an economic program that sent shock waves through the markets and divided her Conservative	2022-10-21 06:47:09
海外ニュース	Japan Times latest articles	Liz Truss defied the markets, and they ruthlessly sealed her fate	https://www.japantimes.co.jp/news/2022/10/21/world/politics-diplomacy-world/liz-truss-britain-resigns-analysis/	truss	2022-10-21 06:41:13
ニュース	BBC News - Home	Liz Truss resigns: Race kicks off to find her successor within a week	https://www.bbc.co.uk/news/uk-politics-63338171?at_medium=RSS&at_campaign=KARANGA	michael	2022-10-20 21:36:15
ニュース	BBC News - Home	Ukraine war: Iranian drone experts 'on the ground' in Crimea - US	https://www.bbc.co.uk/news/world-europe-63329266?at_medium=RSS&at_campaign=KARANGA	house	2022-10-20 21:39:51
ニュース	BBC News - Home	Chess cheating row: Hans Niemann sues accusers Magnus Carlsen and Chess.com for libel	https://www.bbc.co.uk/news/world-us-canada-63338375?at_medium=RSS&at_campaign=KARANGA	carlsen	2022-10-20 21:25:13
ニュース	BBC News - Home	Liz Truss: Six moments from the PM's six chaotic weeks	https://www.bbc.co.uk/news/uk-politics-63334457?at_medium=RSS&at_campaign=KARANGA	dramatic	2022-10-20 21:32:50
ニュース	BBC News - Home	Leicester City 2-0 Leeds United: Foxes claim valuable win against fellow strugglers to climb off the bottom	https://www.bbc.co.uk/sport/football/63238010?at_medium=RSS&at_campaign=KARANGA	Leicester City Leeds United Foxes claim valuable win against fellow strugglers to climb off the bottomManager Jesse Marsch is booed by Leeds fans as Leicester City secure a valuable Premier League victory to move off the bottom of the table	2022-10-20 21:45:49
ニュース	BBC News - Home	PSG 0-1 Chelsea: Millie Bright goal gets visitors off to winning start in Champions League	https://www.bbc.co.uk/sport/football/63321387?at_medium=RSS&at_campaign=KARANGA	PSG Chelsea Millie Bright goal gets visitors off to winning start in Champions LeagueMillie Bright s first half volley earns Chelsea victory over Paris St Germain in their opening Champions League Group A game	2022-10-20 21:29:09
ニュース	BBC News - Home	Fulham 3-0 Aston Villa: Steven Gerrard 'not going to quit' as pressure grows	https://www.bbc.co.uk/sport/av/football/63339294?at_medium=RSS&at_campaign=KARANGA	Fulham Aston Villa Steven Gerrard x not going to quit x as pressure growsUnder pressure Steven Gerrard says he won t quit as Aston Villa boss after their poor form continued with a defeat to Fulham	2022-10-20 21:31:08
ビジネス	ダイヤモンド・オンライン - 新着記事	米スナップ増収率鈍化続く　上場来最低に - WSJ発	https://diamond.jp/articles/-/311703	鈍化	2022-10-21 06:24:00
ビジネス	東洋経済オンライン	話が致命的にわかりにくい人は｢順序｣を知らない ｢結論⇒理由⇒事例⇒結論｣で組み立ててみよう | リーダーシップ・教養・資格・スキル | 東洋経済オンライン	https://toyokeizai.net/articles/-/626851?utm_source=rss&utm_medium=http&utm_campaign=link_back	東洋経済オンライン	2022-10-21 06:30:00