投稿時間:2021-10-27 09:47:28 RSSフィード2021-10-27 09:00 分まとめ(57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	ITmedia 総合記事一覧	[ITmedia ビジネスオンライン] 日本電産・永守会長「小型EV向け注力」	https://www.itmedia.co.jp/business/articles/2110/27/news077.html	itmedia	2021-10-27 08:28:00
IT	ITmedia 総合記事一覧	[ITmedia エグゼクティブ] 日本電産・永守会長「小型EV向け注力」	https://mag.executive.itmedia.co.jp/executive/articles/2110/27/news076.html	itmedia	2021-10-27 08:07:00
TECH	Techable（テッカブル）	文化財をデータとして後世に残す！ 彫刻職人による「つなぎの龍」を1/5サイズのレプリカにして秩父神社に奉納	https://techable.jp/archives/165199	左甚五郎	2021-10-26 23:00:56
python	Pythonタグが付けられた新着投稿 - Qiita	rospy.loginfo関数を使ってPrint	https://qiita.com/hoshianaaa/items/5c5738e85221a86efa90		2021-10-27 08:13:03
js	JavaScriptタグが付けられた新着投稿 - Qiita	【備忘録】Ajax について勉強したのでまとめてみる: XMLHttpRequest	https://qiita.com/machinoyumi/items/35bbc771b2f2889507b4	【備忘録】Ajaxについて勉強したのでまとめてみるXMLHttpRequestはじめにAJAXとは何AJAXとはプログラミング言語ではなく、AsynchronousJavaScriptANDXMlの略。	2021-10-27 09:00:14
Program	[全てのタグ]の新着質問一覧｜teratail（テラテイル）	JS 正規表現で置換	https://teratail.com/questions/366389?rss=all		2021-10-27 08:42:48
Program	[全てのタグ]の新着質問一覧｜teratail（テラテイル）	VBAを使いpdf操作	https://teratail.com/questions/366388?rss=all	VBAを使いpdf操作お世話になります。	2021-10-27 08:10:54
Program	[全てのタグ]の新着質問一覧｜teratail（テラテイル）	Rails：ホットペッパーグルメAPIを用いて現在地に基づくお店検索機能をつけたい	https://teratail.com/questions/366387?rss=all	RailsホットペッパーグルメAPIを用いて現在地に基づくお店検索機能をつけたい前提・実現したいことステップアップのため、学習としてWebAPIを使ったアプリを開発したく、ホットペッパーグルメAPIを用いて条件ごとにレストラン情報を検索するアプリを自作しようと考えています。	2021-10-27 08:03:26
Ruby	Rubyタグが付けられた新着投稿 - Qiita	【Ruby on Rails】簡単なAPI作ってみた	https://qiita.com/mokio/items/7d36a7aa79a4bf5066e9	APIモードで作成するとAPIには不要ないViewなどがインストールされません。	2021-10-27 08:13:41
Git	Gitタグが付けられた新着投稿 - Qiita	Gitでトラブった際のコマンド	https://qiita.com/joe-main/items/381ebd66d4766945cfd5	ブランチ名を間違えたgitbranchmブランチ名急な緊急対応で一時的に作業を差し込むパターンその時点で作業が完了していればコミットするgitcommitmメッセージパターンコミットできなければ一時避難させるgitstashブランチを切り替えて作業、完了後に元に戻すgitstashpop誤ったコミットをリモートにプッシュした時誤ったブランチを削除ex自分がプッシュしたfeatureブランチ等gitpushoriginブランチ名自分のローカルのコミットを元に戻すgitresethardコミットID改めて再度プッシュするgitpushoriginブランチ名リリースしてバグがあったのですぐ切り戻しを行うコミットログを確認してマージコミットを見つけるgitlogマージコミットを指定して打ち消しコミットを作るgitrevertmマージコミットIDこのmはmainlineの略でparentnumberを指定。	2021-10-27 08:01:52
Ruby	Railsタグが付けられた新着投稿 - Qiita	【Ruby on Rails】簡単なAPI作ってみた	https://qiita.com/mokio/items/7d36a7aa79a4bf5066e9	APIモードで作成するとAPIには不要ないViewなどがインストールされません。	2021-10-27 08:13:41
技術ブログ	Developers.IO	QuickSightでSPICEの増分更新が出来るようになりました	https://dev.classmethod.jp/articles/quicksight-spice-incremental-refresh/	quicksight	2021-10-26 23:29:47
技術ブログ	Developers.IO	別アカウントの EC2 インスタンスにクレデンシャルをセットしてクロスアカウントでないアクセスをしてみた	https://dev.classmethod.jp/articles/not-cross-account-access/	構成	2021-10-26 23:22:26
海外TECH	Ars Technica	Promising-looking SETI signal turns out to be of human origin	https://arstechnica.com/?p=1807825	researchers	2021-10-26 23:14:19
海外TECH	DEV Community	Q&A: How can AI be used in banking?	https://dev.to/mage_ai/qa-how-can-ai-be-used-in-banking-16p5	Q amp A How can AI be used in banking TLDRThe financial services or banking industry is an essential part of our everyday lives but the institutions who adopt and integrate artificial intelligence AI will have a clear advantage for their future business success OutlineBanking as you know itAI use cases todayCase studiesConclusion Banking as you know itTraditionally banks provided consumers a safe and secure method of saving and storing their money credit to buy large purchases such as homes and automobiles and other services such as wealth management Though the general purpose of banks and financial institutions have remained the same the way we “bank has changed significantly within the last few decades With the rise of telephone and internet banking in the s and s and now with the disruption of fintechs we ve gone from going to a brick and mortar institution or ATM to “pull out cash to a more cashless society of peer to peer pp payments such as Venmo PayPal Zelle or Cash App We can t forget contactless payments such as Apple Pay Google Pay and Samsung Pay that may have you wondering if we even need banks at all These drastic changes came about with the investment in technology and the ever increasing amount of data From banks to credit unions to fintechs they were able to leverage data to improve customer experience making the process of banking easier and more personalized According to a recent Deloitte survey of financial services AI adopters say that AI will be very or critically important to their business s success in the next couple of years AI has already transformed the banking industry globally in a short span of time We ll take a look at some ways AI has impacted and shifted the financial services industry AI use cases today Fraud detectionYou re enrolled in it You could ve had someone pose as you today and not even know because AI is protecting you AI is helping with the fight against money related fraud and scams through fraud detection It follows the steps of detection investigation and then “dealing with it The process stems from an unusual pattern of the payment transaction Applying AI on your behavioral patterns determines whether the payment is legitimate or not taking into account frequency of purchase and location of prior purchases It s not about comparing handwriting or signatures especially in the world of contact less pay In the case that it is flagged as potentially fraud the bank launches an “investigation by texting the phone number of the owner s account Then the owner gets a notification to verify whether the transaction is real or illegitimate Only trust verified application notifications not text messages Source BofA LendingAt some point in our lives we may decide to make some large purchases such as buying a car or home It s not uncommon to borrow from a bank to make such purchases under the right terms and circumstances If you ve ever applied for a loan you know how nerve wracking it could be or maybe even frustrating In the past the manual verification process of lenders to check transaction history credit scores and other factors could take many hours if not days to get your approval AI based credit decision systems today can analyze consumer transaction data and determine eligibility for the loan in the matter of minutes Can I get an AI men Source Meme Generator Furthermore AI is reducing the potential for human error and bias in underwriting and loan origination Bad underwriting was a huge factor in the recession AI companies like underwrite ai apply machine learning to radically outperform traditional scorecards in both consumer and small business lending while mitigating human errors Risk managementIt is common to have an actuary at a bank to handle risk management In other words determine how to calculate insurance prices and premiums The occupation generalizes a person s history behavior and other personal private information to forecast the likelihood of what will happen to them in the future In your future I see…AISimilarly AI can do risk management too Companies can employ neutral networks to explore an infinite realm of possibilities given the client s personal private information Conversely actuaries can use their knowledge to train regression models that make predictions focusing on forecasting Customer ServiceNo one enjoys calling the bank to dispute a claim or a credit hold but everyone has to do it eventually Have you ever sat down listening to a phone tree It ruins the customer experience and is dreadfully slow and emotionally taxing We ve all been there Source Giphy Banks have multiple legal steps to get consent before processing This is where AI can shine in banking Using an AI chatbot to replace a phone tree helps connect customers to their goals faster by recommending relevant questions answers and documents Now that s a experience Case studies Does Technology Help or Hurt Morale The Harvard Business Review tried to tackle this question about banks switching to AI The results were as they assumed “no one had time to learn a complicated new system Some people refused to attend the training Others brought their laptops to class and worked the entire time Learning a new technology is challenging and can hurt morale especially if they re not used to the training topics and have multiple learning spikes It s important to develop AI tools that are designed for end users to enjoy with an onboarding process that builds on the basic foundations of what they currently do The ideal onboarding process should have employees excited about learning and improving the existing customer service experience It s as easy as ABC Source Giphy Shift to No Signature PurchasingRecently banks made the decision to remove responsibility for signatures No one reads them or verifies them with the back of the card and most customers scribble whatever they like In the case study the conclusion for this decision was because “security measures and fraud protection continue to improve making your signature unnecessary This is a huge benefit for AI by doing what it should be doing removing mindless tasks that most don t want to do with often a low return on investment Time spent having a human verify signatures is extra time for people to wait to get to their turn in the line Almost there… Source Giphy ConclusionWhether you currently work for a financial institution or simply keep your money in one you know that technology has made the experience on both sides so much better AI has already made a big impact customer service fraud prevention risk management automation etc in banking and will have a greater presence in the industry in the near future Are you wanting to get started with AI but not sure where to start Contact us to see if Mage is a good fit for you Co written by Nathaniel Tjandra and Thomas Chung	2021-10-26 23:39:51
Apple	AppleInsider - Frontpage News	Hands on: Should you buy the Nike or standard aluminum Apple Watch Series 7	https://appleinsider.com/articles/21/10/26/hands-on-should-you-buy-the-nike-or-standard-aluminum-apple-watch-series-7?utm_medium=rss	Hands on Should you buy the Nike or standard aluminum Apple Watch Series If you re pondering a new Apple Watch Series purchase and are torn between the Nike and standard aluminum versions check out this hands on with a breakout of all the differences and our recommendation Nike Apple Watch Series and the green aluminum Apple Watch Series Here s the important bit ーthere are no feature differences between the Nike and non Nike models You don t get any additional sensors one isn t faster than the other and neither comes with bonus accessories Read more	2021-10-26 23:25:48
海外科学	NYT > Science	F.D.A. Panel Recommends Vaccine for Children 5 to 11	https://www.nytimes.com/2021/10/26/us/fda-pfizer-covid-vaccine-kids.html	authorization	2021-10-26 23:29:49
海外科学	NYT > Science	Your Questions About Covid Vaccine Dosage for Children, Answered	https://www.nytimes.com/2021/10/26/well/live/child-vaccine-dose-weight-age.html	answeredwe	2021-10-26 23:42:34
海外科学	NYT > Science	Biden’s Plan to Vaccinate Young Children 5 to 11	https://www.nytimes.com/2021/10/20/us/politics/kids-covid-vaccination.html	Biden s Plan to Vaccinate Young Children to White House officials anticipating the approval of coronavirus shots for to year olds within weeks will rely on doctors clinics and pharmacies instead of mass inoculation sites	2021-10-26 23:43:28
金融	金融総合：経済レポート一覧	グリーン金融政策に銀行はどう対応すればよいのか	http://www3.keizaireport.com/report.php/RID/472794/?rss	大和総研	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	いざ円高になると責任を押し付けられてしまう日銀～日銀は円安を静観へ：Market Flash	http://www3.keizaireport.com/report.php/RID/472795/?rss	marketflash	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	金融政策の正常化は世界の住宅バブル崩壊を引き起こすか：木内登英のGlobal Economy ＆ Policy Insight	http://www3.keizaireport.com/report.php/RID/472797/?rss	lobaleconomypolicyinsight	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	FX Daily(10月25日)～ドル円、113円台後半で方向感乏しく推移	http://www3.keizaireport.com/report.php/RID/472799/?rss	fxdaily	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	地域金融機関による不動産の有効活用～発注者支援の手法も活用した川崎信用金庫の事例：金融調査情報	http://www3.keizaireport.com/report.php/RID/472810/?rss	中小企業	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	「波平さんモデル」から「人生100年モデル」へ～金融大動脈は企業金融から世代間金融・資産運用に転換：高田レポート	http://www3.keizaireport.com/report.php/RID/472812/?rss	岡三証券	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	日経平均先物とTOPIX先物のハミルトニアン・モンテカルロ法による安定分布のベイズ推定（第1回）：先物・オプションレポート 2021年10月号	http://www3.keizaireport.com/report.php/RID/472814/?rss	安定分布	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	リスクアペタイト・フレームワーク（RAF）の現状2021～RAFの高度化の余地、とりわけ「地方銀行・第二地方銀行」に：金融規制（バーゼル規制その他）	http://www3.keizaireport.com/report.php/RID/472830/?rss	地方銀行	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	2021年第2四半期のIPO：世界全体のIPO活動が記録的ペースで続く	http://www3.keizaireport.com/report.php/RID/472832/?rss	eyjapan	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	共通ポイント・ハウスポイントに関する法人アンケート調査を実施（2021年） 大企業連携型共通【概要】～大企業連携型共通ポイントの導入事業者（小売業・飲食業・サービス業）では今後も現状のポイント施策を維持する傾向がみられ、追加導入検討は一部	http://www3.keizaireport.com/report.php/RID/472833/?rss	矢野経済研究所	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	米国企業の高収益体質はコロナ禍でより強化された：鳥瞰の眼・虫瞰の眼	http://www3.keizaireport.com/report.php/RID/472835/?rss	鳥瞰	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	IMF国際金融安定化報告書（10月）の注目点～脆弱性の高まりに警鐘 政策支援とリスク抑制の両立を：マーケット・レポート	http://www3.keizaireport.com/report.php/RID/472836/?rss	高まり	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	【石黒英之のMarket Navi】脱炭素と日本株を考える～投資信託を通じて脱炭素に...	http://www3.keizaireport.com/report.php/RID/472837/?rss	marketnavi	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	オーストラリア マーケット動向（2021/10/26）～先週の豪ドルの対円レートは、多少上昇	http://www3.keizaireport.com/report.php/RID/472838/?rss	三井住友	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	2021年11月の注目イベント～日本の新政権の勢力が判明、米国はテーパリング開始となるか	http://www3.keizaireport.com/report.php/RID/472839/?rss	三井住友	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	「ユニコーン新潮流」（2）「その他」の台頭、フィンテックが牽引～ブロックチェーンも地域分散を加速：JCER 中国・アジアウォッチ	http://www3.keizaireport.com/report.php/RID/472852/?rss	日本経済研究センター	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	ソーシャルボンドガイドライン 2021年10月策定（和文・英語版）	http://www3.keizaireport.com/report.php/RID/472870/?rss	金融庁	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	DC運用商品の除外方法の拡充～運用商品の洗練化が加速する	http://www3.keizaireport.com/report.php/RID/472881/?rss	除外	2021-10-27 00:00:00
金融	金融総合：経済レポート一覧	【ミャンマー】輸出外貨管理規制と為替動向について：Mizuho Country Focus	http://www3.keizaireport.com/report.php/RID/472882/?rss	mizuhocountryfocus	2021-10-27 00:00:00
金融	日本銀行：RSS	預金種類別店頭表示金利の平均年利率等	http://www.boj.or.jp/statistics/dl/depo/tento/te211027.pdf	預金	2021-10-27 08:50:00
ニュース	BBC News - Home	Wage rises will put 30p on a pint, says pub chain	https://www.bbc.co.uk/news/business-59001720?at_medium=RSS&at_campaign=KARANGA	minimum	2021-10-26 23:03:55
ニュース	BBC News - Home	Covid: Charlize Theron wants fairer distribution of vaccines	https://www.bbc.co.uk/news/world-africa-59054894?at_medium=RSS&at_campaign=KARANGA	goals	2021-10-26 23:32:36
ニュース	BBC News - Home	Climate activists hold Science Museum protest over fossil fuel sponsors	https://www.bbc.co.uk/news/uk-england-london-59055942?at_medium=RSS&at_campaign=KARANGA	sponsors	2021-10-26 23:49:12
ニュース	BBC News - Home	Life at 50C: Surviving in Kuwait's 'unbearable' heat	https://www.bbc.co.uk/news/world-middle-east-59054893?at_medium=RSS&at_campaign=KARANGA	everyday	2021-10-26 23:10:35
ニュース	BBC News - Home	Julian Assange: WikiLeaks founder's extradition appeal explained	https://www.bbc.co.uk/news/uk-59053803?at_medium=RSS&at_campaign=KARANGA	wikileaks	2021-10-26 23:02:18
ニュース	BBC News - Home	Ros Atkins on... India's climate challenge	https://www.bbc.co.uk/news/world-asia-india-59049691?at_medium=RSS&at_campaign=KARANGA	commitments	2021-10-26 23:08:41
ニュース	BBC News - Home	Girls Night In: 'Spiking is part of going out - so we're staying in'	https://www.bbc.co.uk/news/uk-59054772?at_medium=RSS&at_campaign=KARANGA	venues	2021-10-26 23:05:15
ニュース	BBC News - Home	Who is the redhead living in the Tower of London?	https://www.bbc.co.uk/news/uk-england-london-58979105?at_medium=RSS&at_campaign=KARANGA	blood	2021-10-26 23:36:00
ニュース	BBC News - Home	Malkhan Singh: The surrender of India's bandit king	https://www.bbc.co.uk/news/world-asia-india-58831394?at_medium=RSS&at_campaign=KARANGA	malkhan	2021-10-26 23:29:59
ニュース	BBC News - Home	Black History Month: Lutalo Muhammad - Why I’m proud to be black and British	https://www.bbc.co.uk/sport/59038186?at_medium=RSS&at_campaign=KARANGA	black	2021-10-26 23:02:25
ニュース	BBC News - Home	Alex Clare: The singer who chose his faith over Adele tour	https://www.bbc.co.uk/news/entertainment-arts-59007454?at_medium=RSS&at_campaign=KARANGA	industry	2021-10-26 23:39:08
ビジネス	ダイヤモンド・オンライン - 新着記事	テーパリング迫るFRB、インフレ見通し揺らぐ - WSJ発	https://diamond.jp/articles/-/285882	見通し	2021-10-27 08:08:00
ビジネス	ダイヤモンド・オンライン - 新着記事	ファイザー製ワクチンの子ども向け接種、米FDA諮問委が承認を勧告 - WSJ発	https://diamond.jp/articles/-/285883	諮問	2021-10-27 08:03:00
LifeHuck	ライフハッカー［日本版］	Apple Watchで「ワークアウト」データの精度を高めるには	https://www.lifehacker.jp/2021/10/244586you-could-be-getting-more-accurate-workout-data-from-yo.html	applewatch	2021-10-27 08:30:00
ニュース	THE BRIDGE	「M＆Aクラウド」運営、STRIVEらから10億円を調達——買い手から売り手へ、熱意が伝わる仕組みを開発	http://feedproxy.google.com/~r/SdJapan/~3/pWrPPsHPw3Y/macloud-jpy1b-funding	「MAクラウド」運営、STRIVEらから億円を調達ー買い手から売り手へ、熱意が伝わる仕組みを開発オンラインMAマッチングプラットフォーム「MAクラウド」を運営するMAクラウドは日、直近のラウンドで約億円を調達したと発表した。	2021-10-27 00:00:05
ニュース	THE BRIDGE	ついにNFTは戦隊モノにまで拡大、玩具大手Hasbroが初のNFTコレクションを販売	http://feedproxy.google.com/~r/SdJapan/~3/C8v9F438YPY/hasbro-launches-first-nft-collection-with-power-rangers-on-wax-blockchain	ついにNFTは戦隊モノにまで拡大、玩具大手Hasbroが初のNFTコレクションを販売HasbroはWAXブロックチェーン上のPowerRangers訳注日本の戦隊モノを発売することで、ノンファンジブル・トークンNFTコレクティブル市場に参入する。	2021-10-26 23:15:34
ニュース	THE BRIDGE	3.5億人のZ世代が利用、eSports「FaZe Clan」がSPAC上場へ	http://feedproxy.google.com/~r/SdJapan/~3/7L0YycWkaek/faze-clan-will-go-public-in-a-1b-spac-as-esports-grows	億人のZ世代が利用、eSports「FaZeClan」がSPAC上場へFaZeClanはeSports、グッズ販売、ソーシャルコンテンツの組み合わせを加速させるために、評価額億ドルの特別目的買収会社SPACを通じて株式公開すると発表した。	2021-10-26 23:00:48
GCP	Cloud Blog	How Pokémon GO scales to millions of requests?	https://cloud.google.com/blog/topics/developers-practitioners/how-pok%C3%A9mon-go-scales-millions-requests/	How Pokémon GO scales to millions of requests Have you caught Pokémons Pokémon GO is a popular game played by millions but it scales extremely well This blog is a behind the scenes look into how the Pokémon GO engineering team manages and maintains the scale Joining me is James Prompanya Senior Engineering Manager at Niantic Labs who leads the server infrastructure team for Pokémon GO Let s see what he had to say when I asked him about the architecture that powers this extremely popular game Checkout the video Priyanka What is Pokémon GO James  It s not your typical mobile game It s a game that involves walking around to catch these little Pokémon creatures that are appearing all around you in the real world It encourages you to go outside explore and discover things using augmented reality A big part of that is the community aspect of it When the game first came out we hadn t built community features into the game yet but players still met with others in real life played together and pointed out when rare powerful Pokémon would appear Everyone sees the same Pokémon and shares the same virtual world so when someone points out a Pokémon you d just see crowds of people running out after it Nowadays we make this a major part of the game by hosting regular live events such as community days raid hours all culminating in GO Fest our annual celebration during the summer and our biggest event of the year During these events transactions go from K per second to close to a million in a matter of minutes as soon as regions come online Priyanka How does the Pokémon GO backend scale to handle peaks in traffic during events such as Pokémon GO Fest  James There are lots of services we scale but Google Kubernetes Engine and Cloud Spannerare the main ones Our front end service is hosted on GKE and it s pretty easy to scale the nodes there ーGoogle Cloud provides us with all the tools we need to manage our Kubernetes cluster The Google Cloud console is easy to use with detailed monitoring graphs tools and logging available to use with just a few clicks The support we get from Google engineers is top notch and they re always able to assist at any given moment or in preparation for our large events such as Pokémon GO Fest  We had Google engineers virtually sitting side by side with us ready to tackle any issues from running such a large scale event it was like having an extra support team working directly with us At any given time we have about Spanner nodes handling traffic We also have thousands of Kubernetes nodes running specifically for Pokémon GO plus the GKE nodes running the various microservices that help augment the game experience All of them work together to support millions of players playing all across the world at a given moment And unlike other massively multiplayer online games all of our players share a single “realm so they can always interact with one another and share the same game state    Priyanka Were you always using Spanner Or did you decide to make that architectural decision as the game got popular  James We started off using Google Datastore It was an easy way to get started without having to worry about managing another piece of infrastructure As the game matured we decided we needed more control over the size and scale of the database We also like the consistent indexing that Cloud Spanner provides which allows us to use more complex database schemas with primary and secondary keys  Finally Datastore is non relational with Atomic amp Durable transactions but we needed a relational database with full consistency Spanner provides all of this plus global ACID transactions Priyanka Let s say I am a player playing the game right now I opened the app to catch Pokémon What is happening behind the scenes how does the request flow work James When a user catches a Pokémon we receive that request via Cloud Load Balancing All static media which is stored in Cloud Storage is downloaded to the phone on the first start of the app  We also have Cloud CDN enabled at Cloud Load Balancing level to cache and serve this content First the traffic from the user s phone reaches Global Load Balancer which then sends the request to our NGINX reverse proxy The reverse proxy then sends this traffic to our front end game service The third pod in the cluster is the Spatial Query Backend This service keeps a cache that is sharded by location This cache and service then decides which Pokémon is shown on the map what gyms and PokéStops are around you the time zone you re in and basically any other feature that is location based The way I like to think about it is the frontend manages the player and their interaction with the game while the spatial query backend handles the map  The front end retrieves information from spatial query backend jobs to send back to the user Priyanka What happens when I hunt a Pokémon down and catch it James When you catch the Pokémon we send an event from the GKE frontend to Spanner via the API and when that write request from the frontend to spanner is complete When you do something to update the map like gyms and PokéStops that request sends a cache update and is forwarded to the spatial query backend  Spanner is eventually consistent once the update is received the spatial data is updated in memory and then used to serve future requests from the frontend Then the frontend retrieves information from the spatial query backend and sends it back to the user We also write the protobuf representation of each user action into Bigtable for logging and tracking data with strict retention policies We also publish the message from the frontend to a Pub Sub topic that is used for the analysis pipeline  Priyanka How do you ensure that two people in the same geographic region see the same Pokémon data and keep that relatively in sync Especially for events James It s actually pretty interesting Everything on our servers is deterministic Therefore even if multiple players are on different machines but in the same physical location all the inputs would be the same and the same Pokémon would be returned to both users There s a lot of caching and timing involved however particularly for events It s very important that all the servers are in sync with settings changes and event timings in order for all of our players to feel like they are part of a shared world Priyanka A massive amount of data must be generated during the game How does the data analytics pipeline work and what are you analyzing James You are correct TB of data per day gets generated and we store all of it in BigQuery and BigTable These game events are of interest to our data science team to analyze player behavior verify features like making sure the distribution of pokemon matches what we expect for a given event marketing reports etc We use BigQuery it scales and is fully managed we can focus on analysis and build complex queries without worrying too much about the structure of the data or schema of the table Any field we want to query against is indexed in a way that allows us to build all sorts of dashboards reports and graphs that we share across the team We use Dataflow as our data processing engine so we run a Dataflow batch job to process the player logs stored in Bigtable  We also have some streaming jobs for cheat detection looking for and responding to improper player signals Also for setting up Pokétops and gyms and habitat information all over the world we take in information from various sources like OpenStreetMap the US Geological Survey and WayFarer where we crowdsource our POI data and combine them together to build a living map of the world Priyanka As the events grow and the traffic grows to millions of users per second how does this system scale  James Yes With the increase in transactions there is an increase in the load throughout the system like data pipeline pub sub BigQuery Streaming and more The only thing that the Niantic SRE team needs to ensure is that they have the right quota for these events and since these are managed services there is much less operational overhead for the Niantic team Priyanka With that much traffic the health of the system is critical How do you monitor the health of their system during these massive events James We use Google Cloud Monitoring which comes built in to search through logs build dashboards and fire an alert if something goes critical The logs and dashboards are very extensive and we are able to monitor various aspects and health of the game in real time Next up James and the Pokémon GO engineering team plan to explore managed Agones Game Servers stay tuned and checkout our entire customer architecture playlist We just took a behind the scenes tour into Pokémon GO s architecture How they use GKE and Spanner for scaling to those peaks and how their data science team works with BigQuery BigTable Dataflow amp Pub Sub for data analytics  What did you think about this story Tell me more about it on Twitter pvergadia  Related ArticleUnder the hood Distributed joins in Cloud SpannerHow do you join two tables when both of them are divided into multiple splits managed by multiple different machines In this blog entry Read Article	2021-10-26 23:15:00