Archives
-
Want to help save the planet? Here’s how to truly value nature.
On Point
by "McKinsey On Point" <publishing@email.mckinsey.com> - 12:29 - 15 Sep 2023 -
Keep Your Transport Vehicles Safe and Efficient with Tire Pressure Monitoring Systems
Keep Your Transport Vehicles Safe and Efficient with Tire Pressure Monitoring Systems
Solutions with TPMS
Provided real-time tire pressure and temperature data.
Instant alerts for tire pressure and temperature deviations from preset thresholds.
Historical tire data for informed maintenance decisions.
.png?width=1100&upscale=true&name=Group%20350451687%20(1).png)
Results
%201-1.png)
Improved system reliability leads to reduced downtime.
Implementing the system results in lower fuel costs by contributing to better fuel efficiency.
Improper tire pressure reduces fuel efficiency.
Uffizio Technologies Pvt. Ltd., 4th Floor, Metropolis, Opp. S.T Workshop, Valsad, Gujarat, 396001, India
by "Sunny Thakur" <sunny.thakur@uffizio.com> - 08:00 - 14 Sep 2023 -
Why is Kafka so fast? How does it work?
Why is Kafka so fast? How does it work?
View in browser This is a sneak peek of today’s paid newsletter for our premium subscribers. Get access to this issue and all future issues - by subscribing today.
Latest articles
If you’re not a subscriber, here’s what you missed this month.
To receive all the full articles and support ByteByteGo, consider subscribing:
With data streaming into enterprises at an exponential rate, a robust and high-performing messaging system is crucial. Apache Kafka has emerged as a popular choice for its speed and scalability - but what exactly makes it so fast?
In this issue, we'll explore:
Kafka's architecture and its core components like producer, brokers, and consumers
How Kafka optimizes data storage and replication
The optimizations that enable Kafka’s impressive throughput and low latency
Let’s dive into Kafka’s core components first.
Kafka Architecture Distilled
In a typical scenario where Kafka is used as a pub-sub messaging middleware, there are 3 important components: producer, broker, and consumer. The producer is the message sender, and the consumer is the message receiver. The broker is usually deployed in a cluster mode, which handles incoming messages and writes them to the broker partitions, allowing consumers to read from them.
Note that Kafka is positioned as an event streaming platform, so the term “message”, which is often used in message queues, is not used in Kafka. We call it an “event”.
The diagram below puts together a detailed view of Kafka’s architecture and client API structure. We can see that although the producer, consumer, and broker are still key to the architecture, it takes more to build a high-throughput, low-latency Kafka. Let’s go through the components one by one.
From a high-level point of view, there are two layers in the architecture: the compute layer and the storage layer.
The Compute Layer
The compute layer, or the processing layer, allows various applications to communicate with Kafka brokers via APIs.
The producers use the producer API. If external systems like databases want to talk to Kafka, it also provides Kafka Connect as integration APIs.
The consumers talk to the broker via consumer API. In order to route events to other data sinks, like a search engine or database, we can use Kafka Connect API. Additionally, consumers can perform streaming processing with Kafka Streams API. If we deal with an unbounded stream of records, we can create a KStream. The code snippet below creates a KStream for the topic “orders” with Serdes (Serializers and Deserializers) for key and value. If we just need the latest status from a changelog, we can create a KTable to maintain the status. Kafka Streams allows us to perform aggregation, filtering, grouping, and joining on event streams.
final KStreamBuilder builder = new KStreamBuilder();final KStream<String, OrderEvent> orderEvents = builder.stream(Serdes.String(), orderEventSerde, "orders");While Kafka Streams API works fine for Java applications, sometimes we might want to deploy a pure streaming processing job without embedding it into an application. Then we can use ksqlDB, a database cluster optimized for stream processing. It also provides a REST API for us to query the results.
We can see that with various API support in the compute layer, it is quite flexible to chain the operations we want to perform on event streams. For example, we can subscribe to topic “orders”, aggregate the orders based on products, and send the order counts back to Kafka in the topic “ordersByProduct”, which another analytics application can subscribe to and display.
The Storage Layer
This layer is composed of Kafka brokers. Kafka brokers run on a cluster of servers. The data is stored in partitions within different topics. A topic is like a database table, and the partitions in a topic can be distributed across the cluster nodes. Within a partition, events are strictly ordered by their offsets. An offset represents the position of an event within a partition and increases monotonically. The events persisted on brokers are immutable and append-only, even deletion is modeled as a deletion event. So, producers only handle sequential writes, and consumers only read sequentially.
A Kafka broker’s responsibilities include managing partitions, handling reads and writes, and managing replications of partitions. It is designed to be simple and hence easy to scale. We will review the broker architecture in more detail.
Since Kafka brokers are deployed in a cluster mode, there are two necessary components to manage the nodes: the control plan and the data plane.
Control Plane
The control plane manages the metadata of the Kafka cluster. It used to be Zookeeper that managed the controllers: one broker was picked as the controller. Now Kafka uses a new module called KRaft to implement the control plane. A few brokers are selected to be the controllers.
Why was Zookeeper eliminated from the cluster dependency? With Zookeeper, we need to maintain two separate types of systems: one is Zookeeper, and the other is Kafka. With KRaft, we just need to maintain one type of system, which makes the configuration and deployment much easier than before. Additionally, KRaft is more efficient in propagating metadata to brokers.
We won’t discuss the details of the KRaft consensus here. One thing to remember is the metadata caches in the controllers and brokers are synchronized via a special topic in Kafka.
Data Plane
The data plane handles the data replication. The diagram below shows an example. Partition 0 in the topic “orders” has 3 replicas on the 3 brokers. The partition on Broker 1 is the leader, where the current data offset is at 4; the partitions on Broker 2 and 3 are the followers where the offsets are at 2 and 3.
Step 1 - In order to catch up with the leader, Follower 1 issues a FetchRequest with offset 2, and Follower 2 issues a FetchRequest with offset 3.
Step 2 - The leader then sends the data to the two followers accordingly.
Step 3 - Since followers’ requests implicitly confirm the receipts of previously fetched records, the leader then commits the records before offset 2.
Record
Kafka uses the Record class as an abstraction of an event. The unbounded event stream is composed of many Records.
There are 4 parts in a Record：
Timestamp
Key
Value
Headers (optional)
The key is used for enforcing ordering, colocating the data that has the same key, and data retention. The key and value are byte arrays that can be encoded and decoded using serializers and deserializers (serdes).
Broker
We discussed brokers as the storage layer. The data is organized in topics and stored as partitions on the brokers. Now let’s look at how a broker works in detail.
Step 1: The producer sends a request to the broker, which lands in the broker’s socket receive buffer first.
Steps 2 and 3: One of the network threads picks up the request from the socket receive buffer and puts it into the shared request queue. The thread is bound to the particular producer client.
Step 4: Kafka’s I/O thread pool picks up the request from the request queue.
Steps 5 and 6: The I/O thread validates the CRC of the data and appends it to a commit log. The commit log is organized on disk in segments. There are two parts in each segment: the actual data and the index.
Step 7: The producer requests are stashed into a purgatory structure for replication, so the I/O thread can be freed up to pick up the next request.
Step 8: Once a request is replicated, it is removed from the purgatory. A response is generated and put into the response queue.
Steps 9 and 10: The network thread picks up the response from the response queue and sends it to the corresponding socket send buffer. Note that the network thread is bound to a certain client. Only after the response for a request is sent out, will the network thread take another request from the particular client.
Keep reading with a 7-day free trial
Subscribe to
ByteByteGo Newsletterto keep reading this post and get 7 days of free access to the full post archives.A subscription gets you:
An extra deep dive on Thursdays Full archive Many expense it with team's learning budget Like
Comment
Restack
© 2023 ByteByteGo
548 Market Street PMB 72296, San Francisco, CA 94104
Unsubscribe
by "ByteByteGo" <bytebytego@substack.com> - 11:39 - 14 Sep 2023 -
Some US drones may now fly across states. What are the benefits of drone deliveries?
On Point
Brought to you by Liz Hilton Segel, chief client officer and managing partner, global industry practices, & Homayoun Hatami, managing partner, global client capabilities
•
The drone next door. The US is getting closer to the day when delivery by drone hits the mainstream. Three American drone operators, including one company that examines infrastructure, a second that develops communications and navigation systems for crewless aircraft, and a third that specializes in drone delivery, were granted permission to fly their uncrewed aerial vehicles out of range of human sight. That could lead the way for businesses to send packages across the country by drone. [Axios]
•
At-home drone deliveries. In early 2022, drones made more than 2,000 commercial deliveries each day across the globe. With their potential to meet the needs of consumers who want prepared foods and other small packages delivered, in addition to making last-mile business-to-business deliveries (such as moving medical samples to labs), drones could become an important part of the delivery supply chain, McKinsey partner Robin Riedel and colleagues share on the Future Air Mobility blog.
•
Cost-effective and planet friendly. As the industry develops, drones could become as cost-effective for package delivery as other modes of transport, our analysis finds. Drones are also environmentally friendly, with carbon dioxide emissions for a single delivery typically lower than those of electric cars and vans and significantly lower than those of gasoline-powered vehicles. Discover six questions to consider when developing a drone strategy, and visit our series, The Next Normal, to discover the future of parcel delivery.
— Edited by Belinda Yu, editor, Atlanta
This email contains information about McKinsey's research, insights, services, or events. By opening our emails or clicking on links, you agree to our use of cookies and web tracking technology. For more information on how we use and protect your information, please review our privacy policy.
You received this email because you subscribed to the On Point newsletter.
Copyright © 2023 | McKinsey & Company, 3 World Trade Center, 175 Greenwich Street, New York, NY 10007
by "McKinsey On Point" <publishing@email.mckinsey.com> - 12:36 - 14 Sep 2023 -
ตรวจสอบ 7 ความท้าทายและตัวอย่างสำหรับศูนย์ข้อมูล Edge!
Schneider Electric
7 ความท้าทายและกรณีลูกค้าสำหรับ Edge Data CenterDear Abul,
เพื่อรองรับการเติบโตของแอพพลิเคชั่นที่ใช้ข้อมูลมาก ความหน่วงต่ำเป็นพิเศษ ศูนย์ข้อมูล Edge ในพื้นที่จำเป็นต้องปรับขนาดในสภาพแวดล้อมที่หลากหลาย อย่างไรก็ตาม การขยายขนาดใหญ่ของศูนย์ข้อมูล Edge ทำให้ผู้ประกอบการศูนย์ข้อมูลพบกับความท้าทายที่หลากหลายในแง่ของพลังงานและการทำความเย็น ประสิทธิภาพการใช้พลังงาน การจัดการและการบำรุงรักษา
คู่มือนี้ อธิบายวิธีที่สถาปัตยกรรม telco cloud และ IT cloud ทำงานร่วมกันเป็นสถาปัตยกรรมเสริม ในขณะที่ออกแบบให้ตรงตามความยืดหยุ่นของศูนย์ข้อมูลและเป้าหมายด้านประสิทธิภาพ เรานำเสนอไม่ว่าจะบรรจบกับสิ่งใด นอกจากนี้ยังแนะนำกรณีการใช้งานจริงสำหรับการออกแบบและสร้างศูนย์ข้อมูล Edge ในพื้นที่แบบกระจายที่ยั่งยืนและยืดหยุ่นตามขนาด เช่น การเพิ่มการใช้พลังงานและประสิทธิภาพการดำเนินงาน การตระหนักถึงความเป็นกลางทางคาร์บอน และลดของเสียให้เหลือน้อยที่สุด
© 2023 Schneider Electric. All Rights Reserved. Schneider Electric is a trademark and the property of Schneider Electric SE, its subsidiaries and affiliated companies. All other trademarks are the property of their respective owners.
by "Noe Noe OO, Schneider Electric" <Marcom.thailand@se.com> - 09:02 - 13 Sep 2023 -
Shocks are inevitable. A dynamic risk management plan can help companies stay competitive.
On Point
Brought to you by Liz Hilton Segel, chief client officer and managing partner, global industry practices, & Homayoun Hatami, managing partner, global client capabilities
•
Internal and external threats. Business risks can come from outside a company—like cyberattacks, inflation, and geopolitics—or from within—like an information leak or a missed opportunity. While leaders and organizations may have different comfort levels with risk, creating a dynamic and flexible approach to risk management could help organizations stay competitive when shocks occur, according to McKinsey Global Institute director Olivia White and coauthors. Risk controls encompass the ways in which companies can identify threats, limit or eliminate them, and prevent or reduce loss.
•
Understanding uncertainty. Preparing for risks like a pandemic, climate change, or supply chain disruption can be daunting for any organization. But strategies like scenario planning can help executives and their teams think through hypotheticals and fill existing gaps in their capabilities. One benefit of scenarios is that by developing a range of possible outcomes, each backed with a sequence of events, it is possible to expand leaders’ thinking. Explore this edition of McKinsey Explainers for three components to a successful risk management strategy.
— Edited by Gwyn Herbein, editor, Atlanta
This email contains information about McKinsey's research, insights, services, or events. By opening our emails or clicking on links, you agree to our use of cookies and web tracking technology. For more information on how we use and protect your information, please review our privacy policy.
You received this email because you subscribed to the On Point newsletter.
Copyright © 2023 | McKinsey & Company, 3 World Trade Center, 175 Greenwich Street, New York, NY 10007
by "McKinsey On Point" <publishing@email.mckinsey.com> - 12:11 - 13 Sep 2023 -
Keep Your Transport Vehicles Safe and Efficient with Tire Pressure Monitoring Systems
Keep Your Transport Vehicles Safe and Efficient with Tire Pressure Monitoring Systems
Solutions with TPMS
Provided real-time tire pressure and temperature data.
Instant alerts for tire pressure and temperature deviations from preset thresholds.
Historical tire data for informed maintenance decisions.
Results
Improved system reliability leads to reduced downtime.
Implementing the system results in lower fuel costs by contributing to better fuel efficiency.
Improper tire pressure reduces fuel efficiency.
Uffizio Technologies Pvt. Ltd., 4th Floor, Metropolis, Opp. S.T Workshop, Valsad, Gujarat, 396001, India
by "Sunny Thakur" <sunny.thakur@uffizio.com> - 08:00 - 12 Sep 2023 -
API Trends, Insights, Webinars, and More!
SmartBear
Hi Abul,
We're back with our monthly API newsletter. Let's dive into this month's API trends, insights, webinars, and more!
Hot Off the PressStoplight is Joining SmartBear!
SmartBear has acquired Stoplight, a leader in API design and documentation
Stoplight supports three leading open source projects that we are excited to support in addition to our existing open source initiatives: Spectral, Prism, and Elements. Check out our blog to find out what’s coming next with this acquisition.
API Innovative Insights 
Become an expert on the latest updates and trends Want to Learn More?
Check out what else has been happening at SmartBear
WEBINARWhat’s New in SwaggerHubJoin our team as we take you through the latest features in SwaggerHub and demonstrate the advantages of integrating them into your toolkit 
WEBINARSupercharge Your Performance TestingOur experts delve into performance and load testing methodologies that can elevate your application's reliability, scalability, and overall user satisfaction
COMMUNITY EVENTCelebrating 10 Years of Pact OS2023 marks 10 years of Pact Open Source. To celebrate, during the month of October – which we’ve coined Pactober – we’re facilitating community events across the globe to celebrate and acknowledge the journey and look forward to the future of Pact
BLOGDiscover, Test, and Succeed: SwaggerHub Explore Meets SwaggerHub PortalAPI documentation is the cornerstone of any successful API-driven project. We’re thrilled to introduce an exciting new enhancement to our SwaggerHub platform that’s set to take your API documentation to new heights
Best,
SmartBear API Team
P.S. If you found this email helpful, forward it to a friend! If you have ideas of how we can improve next month’s newsletter, reply back to let us know.Have you tried our free API client, Explore? Sign up Today!
by "SmartBear API Team" <api-lifecycle-team@smartbearmail.com> - 02:46 - 12 Sep 2023 -
FutureStack London is back!
New Relic
Our FutureStack user conference is coming to London on Tuesday 14th November and will feature inspiring customer presentations, exciting product updates, and multiple hands-on workshops.
Come and meet Grok, our industry-first new GenAI assistant and learn about what’s new and what’s next from New Relic with exciting announcements around logs, security, infrastructure, unified APM and the developer toolchain.
You’ll hear customer stories of observability excellence from engineering leaders at some of our biggest customers, and get to up-level your observability game with hands-on workshops for both new and advanced users.
So come tap into the expertise of our New Relic engineers while having a little fun to round off the day.
Register today and we look forward to seeing you in November!
Register here 
Need help? Let's get in touch.
This email is sent from an account used for sending messages only. Please do not reply to this email to contact us—we will not get your response.
This email was sent to info@learn.odoo.com Update your email preferences.
For information about our privacy practices, see our Privacy Policy.
Need to contact New Relic? You can chat or call us at +44 20 3859 9190.
Strand Bridge House, 138-142 Strand, London WC2R 1HH
© 2023 New Relic, Inc. All rights reserved. New Relic logo are trademarks of New Relic, Inc
Global unsubscribe page.
by "New Relic EMEA Events" <emeamarketing@newrelic.com> - 09:22 - 12 Sep 2023 -
What would it take to empower Black, Latina, and Native American women in tech?
On Point
Brought to you by Liz Hilton Segel, chief client officer and managing partner, global industry practices, & Homayoun Hatami, managing partner, global client capabilities
•
Diversifying cybersecurity. In America, roughly three in ten cybersecurity jobs go unfilled due to a lack of workers, government data shows. As cyberattacks grow more frequent and costly, organizations are aiming to recruit more women and people of color to the field. Referring to internal diversity programs in job postings and only mentioning the required certifications and skills for a job can help companies recruit cybersecurity professionals more easily, the CEO of a cybersecurity education company says. [Axios]
•
Shrinking representation. Black, Latina, and Native American (BLNA) women in the US earned nearly double the number of computing degrees in 2021 as they did in 2016. Despite this, and even as employers invest significant time and resources in improving representation and inclusion in the workplace, BLNA women’s share of the technical workforce is shrinking. Between 2018 and 2022, BLNA women’s representation in tech roles declined from 4.6% to 4.1%, McKinsey senior partner Tiffany Burns and coauthors explain.
— Edited by Belinda Yu, editor, Atlanta
This email contains information about McKinsey's research, insights, services, or events. By opening our emails or clicking on links, you agree to our use of cookies and web tracking technology. For more information on how we use and protect your information, please review our privacy policy.
You received this email because you subscribed to the On Point newsletter.
Copyright © 2023 | McKinsey & Company, 3 World Trade Center, 175 Greenwich Street, New York, NY 10007
by "McKinsey On Point" <publishing@email.mckinsey.com> - 10:07 - 11 Sep 2023 -
สิทธิพิเศษ! รับการประเมินไซต์งานด้วย EcoConsult จาก ชไนเดอร์ อิเล็คทริค
Schneider Electric
EcoConsult สำหรับธุรกิจของคุณ 
เพิ่มประสิทธิภาพการบำรุงรักษาสินทรัพย์ไฟฟ้าในไซต์งานเรียน คุณ Abul
ในฐานะที่ ชไนเดอร์ อิเล็คทริค เป็นส่วนหนึ่งของการร่วมผลักดันองค์กรและพาร์ทเนอร์ทั่วโลกเข้าสู่ Net Zero เราขอเสนอ การประเมินทรัพย์สินทางไฟฟ้าให้กับเจ้าของไซต์งานและผู้จัดการงานอาคารทั้งหมด โดยไม่มีข้อผูกมัดใดๆ
ปัจจุบันผู้จัดการงานอาคาร ต่างเผชิญกับความท้าทาย ในการรักษาสมดุลระหว่างการบำรุงรักษาไซต์งานให้ดำเนินการได้อย่างราบรื่น โดยใช้ทรัพยากรที่มีให้เกิดประสิทธิภาพสูงสุด ในขณะที่ต้องรับมือกับการบริหารลงทุนที่ต่างมีอยู่อย่างจำกัด เกี่ยวกับการจัดการโครงสร้างพื้นฐานและการเปลี่ยนถ่ายอุปกรณ์ไฟฟ้าต่างๆ
อย่างไรก็ตาม ท่านยังคงมีความจำเป็นในการบริหารไซต์งานให้ทำงานต่อไปด้วยทรัพยากรที่มีอยู่ สู่แนวทาง "Make do with what you have"โชคร้ายที่ข้อจำกัดและความเสี่ยงเหล่านี้มักทำให้ท่านอาจะพบกับตัน จนอาจทำให้การบริการจัดการไซต์งาน/อาคาร จนอาจเกิดความล้มเหลว
หากเกิดการหยุดทำงาน โดยไม่ได้วางแผนมีการวางแผนล่วงหน้า ทำให้ความรับผิดชอบทั้งหมดมักตกอยู่ที่ท่าน ผู้จัดการงานอาคาร! พร้อมแบกรับความผิดชอบและมองหาวิธีแก้ปัญหาตลอดเวลาการประเมินไซต์งาน EcoConsult ของ ชไนเดอร์ อิเล็คทริค เสนอการประเมินสินทรัพย์ทางไฟฟ้า เพื่อช่วยเหลือคุณในรูปแบบต่างๆ:- มองเห็นภาพรวมอุปกรณ์ที่ติดตั้ง: แสดงภาพรวมครอบคลุมอุปกรณ์ที่ใช้งานอยู่ในปัจจุบัน
- การตรวจจับอันตรายที่ใกล้เข้ามา: ระบุความเสี่ยงและอันตรายที่อาจคุกคามการทำงานของไซต์งาน
- รายละเอียดของอุปกรณ์: ดูรายละเอียดของอุปกรณ์เฉพาะที่มีการจัดแบ่งหมวดหมู่ต่างๆ
- การปรับปรุงแผนให้ทันสมัย: ปรับให้ทันสมัย ตามความสำคัญและความต้องการเฉพาะในแต่ละการติดตั้ง โดยมีวัตถุประสงค์ช่วยจัดการกลุ่มอุปกรณ์ที่ล้าสมัย
ข้อดีของรายงานการประเมิน:ดูรายละเอียดจากรายงานที่แนะนำโดยผู้เชี่ยวชาญเพื่อเตรียมความพร้อมที่ดีขึ้น:- วางแผนงบประมาณ: สามารถวางแผนการบริหารทรัพยากรและงบประมาณที่จำเป็นสำหรับการบำรุงรักษาและปรับปรุงไซต์งาน
- ใช้โปรโตคอลการบำรุงรักษา: ดำเนินการทางโปรโตคอลการบำรุงรักษาที่มีโครงสร้างดี ปรับให้เหมาะกับสถานที่การติดตั้ง
- การรีเฟรชอุปกรณ์และชิ้นส่วนอะไหล่: สำรวจและรับการเปลี่ยนถ่ายอุปกรณ์และชิ้นส่วนอะไหล่ที่จำเป็น อย่างมีกลยุทธ์ก่อนที่มันจะล้าสมัย
การตรวจสอบจะถูกดำเนินการอย่างไม่มีข้อผูกมัดใดๆ:ท่านสามารถวางใจและรับการให้คำแนะนำสำหรับไซต์งาน เพื่อประโยชน์อันสูงสุด การดำเนินการทั้งหมดจะไม่มีข้อผูกมัดและไม่ยึดติดกับแบรนด์ทั้งสิ้นสิทธิประโยชน์แก่ท่านในฐานะผู้จัดการไซต์งานและสถานที่ ในการรับคำปรึกษา ตลอดจนขอความคิดเห็นจากผู้เชี่ยวชาญตัวจริง!
ด้วยการดำเนินงานที่ครอบคลุมทั่วโลกในกว่า 100 ประเทศ เราพร้อมนำประสบการณ์อันเป็นประโยชน์มอบแก่ท่าน เพื่อช่วยให้ท่านทราบวิธีจัดการความต้องการ ในการบำรุงรักษาอุปกรณ์ไฟฟ้าและไซต์งานของคุณอย่างมีประสิทธิภาพ ทำให้เกิดประสิทธิผลมากที่สุด
หากท่านสนใจเพียงคลิกที่ปุ่ม 'ติดต่อเรา' และทีมงานของเราจะติดต่อกลับโดยเร็ว
ขอแสดงความนับถือ,วราชัย จตุรสถาพรBusiness Vice PresidentField Services© 2023 Schneider Electric. All Rights Reserved. Schneider Electric is a trademark and the property of Schneider Electric SE, its subsidiaries and affiliated companies. All other trademarks are the property of their respective owners.
by "Schneider Electric" <reply@se.com> - 10:02 - 11 Sep 2023 -
Some employees are destroying value. Others are building it. Do you know the difference?
Share this email
Brought to you by Liz Hilton Segel, chief client officer and managing partner, global industry practices, & Homayoun Hatami, managing partner, global client capabilities
Employee disengagement and attrition could cost a median-size S&P 500 company more than $288 million a year in lost productivity, according to recent McKinsey research. The more satisfied and committed employees are at work, the higher their self-reported performance and well-being. With hybrid and remote-working models now the norm in many workplaces, how can companies measure employee effectiveness, and in turn, boost productivity? In a new McKinsey Quarterly article, Aaron De Smet, Angelika Reich, and their coauthors describe the six worker archetypes present in every organization—from highly dissatisfied and actively disengaged on one end of the spectrum to “thriving stars” on the other—and suggest what companies can do to help improve the performance of all employees. Check it out to see how your company can build a more resilient and engaged workforce. And don’t miss this week’s sustainable and inclusive growth briefing which features the best-of-summer roundup of SIG insights.
Quote of the day
Chart of the day
ALSO NEW
— Edited by Joyce Yoo, editor, New York
Follow our thinking
Share these insights
Did you enjoy this newsletter? Forward it to colleagues and friends so they can subscribe too. Was this issue forwarded to you? Sign up for it and sample our 40+ other free email subscriptions here.
This email contains information about McKinsey’s research, insights, services, or events. By opening our emails or clicking on links, you agree to our use of cookies and web tracking technology. For more information on how we use and protect your information, please review our privacy policy.
You received this email because you subscribed to our McKinsey Quarterly alert list.
Copyright © 2023 | McKinsey & Company, 3 World Trade Center, 175 Greenwich Street, New York, NY 10007
by "McKinsey Daily Read" <publishing@email.mckinsey.com> - 06:23 - 11 Sep 2023 -
Cybersecurity in the age of generative AI: A leader’s guide
Share this email
Brought to you by Liz Hilton Segel, chief client officer and managing partner, global industry practices, & Homayoun Hatami, managing partner, global client capabilities
Cyber Awareness Month begins on October 1 in the United States, but it’s never too early for leaders anywhere to think about cybersecurity—especially now that generative AI (gen AI) has profoundly altered the online landscape. Businesses that are eager to take advantage of gen AI’s considerable benefits cannot ignore its increasing risks. Hallucinations, copyright infringement, and advanced social engineering and impersonation are just a few of the many threats that gen AI technology poses to organizations, most of which are already grappling with an ever-increasing number of cyberattacks. Here’s a quick look at what your organization should be prepared for.
It’s not often that a $2 trillion opportunity comes along. That’s the estimated worth of the market for cybersecurity products and services, according to new research led by McKinsey partner Marc Sorel and colleagues. “Currently available commercial solutions do not fully meet customer demands in terms of automation, pricing, services, and other capabilities,” they say. Only about 10 percent of the addressable market is served today; combined with increasing regulation and growing C-suite concerns about security and privacy, this gives both cybersecurity providers and buyers a compelling opportunity. “Until recently, many organizations that required cyber protection were not fully engaged with the challenges they faced,” say the McKinsey experts. “Often, they saw the cost and complexity of action as greater than the need for it. Now, with attacks becoming more frequent, the risk–benefit equation has changed.”
That’s the percentage of respondents to a McKinsey survey who rank cybersecurity as one of the top risks of gen AI adoption. Only 38 percent of organizations are working to mitigate the risk, however, down from 51 percent in 2022. Overall, respondents’ mitigation efforts lag behind their concerns; for example, 39 percent of organizations consider privacy a risk, but only 20 percent are making efforts to counter it. One way to allay risks is to “keep a human in the loop,” suggest McKinsey experts. “That is, make sure a real human checks any gen AI output before it’s published or used.”
That’s McKinsey’s Bryce Hall quoting “one of the most colorful analogies” on gen AI that he has heard from business leaders. “With the speed and advancements in technology, we don’t even know yet what type of electric fence to put into place,” adds McKinsey expert Liz Grennan. “We are seeing a lot of case studies about reputational damage, customer attrition, and erosion of market value, along with increased fines and regulatory scrutiny.” Leading companies that are developing or adopting gen AI solutions are pulling in legal and cyber risk experts from the start, and some global compliance standards may take effect next year. “The stage will be very crowded with both virtue and vice,” says Grennan. “One of next year’s tech trends could be the use of AI to combat the harms of AI because we can’t solve this with traditional means.”
A strong culture often contributes to the success of most corporate initiatives, and cybersecurity is no exception. Developer data platform company MongoDB discovered this when it implemented a security awareness program that encouraged employees to participate in protecting the organization. Roughly more than 100 of the firm’s employees—people from different ranks and locations and with varying levels of expertise—volunteer as “security champions” to test security-related tools, identify potential vulnerabilities, and provide feedback to the company’s security team. A clear win is when feedback from participants leads to a tool’s adoption, says Felix Chen, MongoDB’s cybersecurity education and advocacy senior analyst, in an interview with McKinsey’s James Kaplan and Charlie Lewis. “We ask them for a simple input. Even if it’s something that we don’t go with, the fact that we are asking creates a sense of inclusion and contribution.”
The Journey to Becoming Enlightened Is Arduous is not, as you might think, a serious work of philosophy—it’s one of many AI-generated fake books on online bestseller lists. And what looks like authentic medical insurance for sale may have a fake human peddling it. While some malicious applications of gen AI technology may be easy to spot, many can be convincing enough to spread political propaganda or distort historical fact. No protection method can be considered foolproof, but organizations that invest in the best talent and build cybersecurity into every process and relationship with customers and suppliers stand a better chance of warding off the newest generation of cyberthreats.
Lead securely.
— Edited by Rama Ramaswami, senior editor, New York
Share these insights
Did you enjoy this newsletter? Forward it to colleagues and friends so they can subscribe too. Was this issue forwarded to you? Sign up for it and sample our 40+ other free email subscriptions here.
This email contains information about McKinsey’s research, insights, services, or events. By opening our emails or clicking on links, you agree to our use of cookies and web tracking technology. For more information on how we use and protect your information, please review our privacy policy.
You received this email because you subscribed to the Leading Off newsletter.
Copyright © 2023 | McKinsey & Company, 3 World Trade Center, 175 Greenwich Street, New York, NY 10007
by "McKinsey Leading Off" <publishing@email.mckinsey.com> - 02:47 - 11 Sep 2023 -
We missed you at How To Build a Smart Queue Management System
We missed you at How To Build a Smart Queue Management System
How to Build a Smart Queue Management System
Watch on-demand 
We're sorry we missed you for our latest DevCon 2023 session How To Build a Smart Queue Management System Step by Step? From Zero to Hero, but not to worry, the on-demand replay is now available for you to watch anytime.
Watch to learn:
- Real-time detection and tracking of people for efficient queue management and staffing optimization
- Step-by-step easy-to-follow Jupyter Notebook tutorial
- Optimized for multi-model workloads across various Intel processors
- Where to find resources; open-source code, dataset, videos, and a blog available on GitHub for easy customization and extension to your specific needs
Watch now 
Don’t Stop There
- Watch tutorials, view use cases and see all the toolkit options at openvino.ai
- Download your free version of Intel® Distribution of OpenVINO™ toolkit and
- Explore the OpenVINO™ toolkit Github repository of Jupyter Notebooks, Training Extensions, Models, and more…
Missed other DevCon sessions? You can watch past workshops on demand as well.
If you forward this email, your contact information will appear in any auto-populated form connected to links in this email.
This was sent to info@learn.odoo.com because you are subscribed to Webinars. To view and manage your marketing-related email preferences with Intel, please click here.
© 2023 Intel Corporation
Intel Corporation, 2200 Mission College Blvd., M/S RNB4-145, Santa Clara, CA 95054 USA. www.intel.com
Privacy | Cookies | *Trademarks | Unsubscribe | Manage Preferences
by "Intel Developer Zone" <intel.developer.zone@plan.intel.com> - 12:31 - 11 Sep 2023 -
Generative AI has entered the mainstream. How widespread is it among workers?
On Point
Brought to you by Liz Hilton Segel, chief client officer and managing partner, global industry practices, & Homayoun Hatami, managing partner, global client capabilities
•
Career builder or breaker? A recent Pew survey finds that workers are split on whether to expect AI technologies to help or hurt their careers over the next 20 years. But IT and technology workers—who are the most exposed to AI—are much more optimistic. They are nearly three times more likely to expect AI to help than to hurt their careers. The Pew research also indicates that women’s roles, which are less likely to require physical labor, are more exposed to AI than men’s. [WaPo]
•
Genuine interest in gen AI. The latest McKinsey Global Survey on AI shows how quickly generative AI tools have caught on. One-third of survey respondents say their organizations are regularly using gen AI in at least one part of their business, and about eight in ten respondents across regions, industries, and seniority levels have at least tried out a gen AI tool. Three-quarters predict that gen AI will cause significant or disruptive change in their industries, and many expect their companies’ AI investments to increase because of advances with gen AI.
•
Expected effects on employees. Senior partners Alexander Sukharevsky, Alex Singla, Lareina Yee and coauthors find with this research that many organizations are not yet fully addressing the risks from gen AI adoption. For example, only 21% of respondents report that their organizations have established policies governing employees’ use of gen AI. Respondents expect gen AI adoption to change their companies’ workforce needs. See which business functions they predict will see the biggest changes and to what extent they expect workers to be reskilled.
— Edited by Heather Hanselman, editor, Atlanta
This email contains information about McKinsey's research, insights, services, or events. By opening our emails or clicking on links, you agree to our use of cookies and web tracking technology. For more information on how we use and protect your information, please review our privacy policy.
You received this email because you subscribed to the On Point newsletter.
Copyright © 2023 | McKinsey & Company, 3 World Trade Center, 175 Greenwich Street, New York, NY 10007
by "McKinsey On Point" <publishing@email.mckinsey.com> - 12:29 - 11 Sep 2023 -
Last chance to register | Leveraging Generative AI to power your Content Supply Chain
Adobe
Adobe Webinar
Leveraging Generative AI to power your Content Supply Chain
Tuesday, 12 September, 2023
11am SGT
Dream it, type it, see it with Firefly, our creative generative AI engine. Now in Photoshop (beta), Illustrator and on the web.
Firefly generative AI capabilities is embedded into other Adobe tools. In this webinar, get a head-start on GenAI with live demos and sharing on the following topics:- Introduction to Generative AI
- Empowering everyone and anyone to express creativity with Adobe Express and Firefly
- How Generative AI fits into your overall content management workflow to drive content at scale
- Enhancing your investment in AEM Assets with Adobe Express and Firefly
Register now
Speakers
Nelson John
Principal Solution Consultant
Adobe
See Wai Yip
Principal Solution Consultant
AdobeCreativity for all.
by "Adobe Creative Cloud for Business" <demand@info.adobe.com> - 09:02 - 10 Sep 2023 -
Meet the partners behind our biggest summer insights
Share this email
Brought to you by Liz Hilton Segel, chief client officer and managing partner, global industry practices, & Homayoun Hatami, managing partner, global client capabilities
New from McKinsey & Company
As summer in the Northern Hemisphere turns to fall, take a moment to get to know some of the partners behind the big articles that made a splash over the summer. These insights share a common theme of preparing for imminent changes in business and society, with generative AI’s promise and potential being one of the biggest topics on the minds of McKinsey leaders, including Sven Blumberg, Dana Maor, Alex Singla, Lareina Yee, Michael Chui, and Bryan Hancock. Get updated on the implications of generative AI, what lies ahead in real estate, economic growth opportunities in Africa, and the ten big shifts facing organizations today.
Lareina Yee
Lareina Yee is a senior partner in the San Francisco Bay Area office and writes and speaks on technology and diversity. She is the chair of the McKinsey Technology Council and also cofounded Women in the Workplace, an annual research partnership with LeanIn.Org, which reports on the state of women in corporate America year over year.
Featured article: What every CEO should know about generative AI
Bryan Hancock
Bryan Hancock is a partner in the Washington, DC, office and the global leader of McKinsey’s talent work. He has served a wide range of talent-intensive businesses, leading employers in sectors such as retail, transportation, logistics, healthcare, banking, asset management, and oil and gas. He is one of the authors of the recent book, Power to the Middle.
To see more essential reading on topics that matter, visit McKinsey Themes.
— Edited by Joyce Yoo, editor, New York
And for insights on issues that matter most to the CEO and their colleagues in the C-suite, sign up for The CEO Shortlist, formerly The Shortlist. We’ve changed the focus (and name) of this newsletter to signal our commitment to helping CEOs, present and future, do the best job they can. Rest assured it will continue to deliver, twice monthly, a shortlist of articles and reports that are must-reads regardless of role—from C-level execs to the frontline.
This email contains information about McKinsey's research, insights, services, or events. By opening our emails or clicking on links, you agree to our use of cookies and web tracking technology. For more information on how we use and protect your information, please review our privacy policy.
You received this email because you subscribed to our McKinsey Global Institute alert list.
Copyright © 2023 | McKinsey & Company, 3 World Trade Center, 175 Greenwich Street, New York, NY 10007
by "McKinsey & Company" <publishing@email.mckinsey.com> - 06:24 - 10 Sep 2023 -
The week in charts
The Week in Charts
Share this email
Share these insights
Did you enjoy this newsletter? Forward it to colleagues and friends so they can subscribe too. Was this issue forwarded to you? Sign up for it and sample our 40+ other free email subscriptions here.
This email contains information about McKinsey's research, insights, services, or events. By opening our emails or clicking on links, you agree to our use of cookies and web tracking technology. For more information on how we use and protect your information, please review our privacy policy.
You received this email because you subscribed to The Week in Charts newsletter.
Copyright © 2023 | McKinsey & Company, 3 World Trade Center, 175 Greenwich Street, New York, NY 10007
by "McKinsey Week in Charts" <publishing@email.mckinsey.com> - 03:38 - 9 Sep 2023 -
EP76: Netflix's Tech Stack
EP76: Netflix's Tech Stack
Forwarded this email? Subscribe here for moreThis week’s system design refresher:
System Design: Apache Kafka In 3 Minutes (Youtube video)
Netflix's Tech Stack
How Do C++, Java, Python Work?
Top 5 Kafka use cases
How is data transmitted between applications?
An Unusual Request: Combating International Book Piracy on Amazon
Engineering Metrics CEOs Love | A Free Presentation Deck
For too many engineering leaders, the most stressful part of their job isn’t a bug or a system crash. The thing they worry about most is having to step into a boardroom and make the case that their engineering team is positively impacting the broader company.
In this CEO-approved slide deck, you’ll find simple ways to communicate how your team is increasing engineering efficiency, all while delivering business results consistently.
From crystal-clear ways to illustrate how engineering resources match company priorities to how developers reduce the turnaround time on essential features, the CTO Board Deck is your secret weapon for owning any boardroom you enter.
System Design: Apache Kafka In 3 Minutes
Netflix's Tech Stack
This post is based on research from many Netflix engineering blogs and open-source projects. If you come across any inaccuracies, please feel free to inform us.
Mobile and web: Netflix has adopted Swift and Kotlin to build native mobile apps. For its web application, it uses React.
Frontend/server communication: GraphQL.
Backend services: Netflix relies on ZUUL, Eureka, the Spring Boot framework, and other technologies.
Databases: Netflix utilizes EV cache, Cassandra, CockroachDB, and other databases.
Messaging/streaming: Netflix employs Apache Kafka and Fink for messaging and streaming purposes.
Video storage: Netflix uses S3 and Open Connect for video storage.
Data processing: Netflix utilizes Flink and Spark for data processing, which is then visualized using Tableau. Redshift is used for processing structured data warehouse information.
CI/CD: Netflix employs various tools such as JIRA, Confluence, PagerDuty, Jenkins, Gradle, Chaos Monkey, Spinnaker, Altas, and more for CI/CD processes.Latest articles
If you’re not a subscriber, here’s what you missed this month.
How Do C++, Java, Python Work?
The diagram shows how the compilation and execution work.
Compiled languages are compiled into machine code by the compiler. The machine code can later be executed directly by the CPU. Examples: C, C++, Go.
A bytecode language like Java, compiles the source code into bytecode first, then the JVM executes the program. Sometimes JIT (Just-In-Time) compiler compiles the source code into machine code to speed up the execution. Examples: Java, C#
Interpreted languages are not compiled. They are interpreted by the interpreter during runtime. Examples: Python, Javascript, Ruby
Compiled languages in general run faster than interpreted languages.
Over to you: which type of language do you prefer?Top 5 Kafka use cases
Kafka was originally built for massive log processing. It retains messages until expiration and lets consumers pull messages at their own pace.
Unlike its predecessors, Kafka is more than a message queue, it is an open-source event streaming platform for various cases.
Let’s review the popular Kafka use cases.Log processing and analysis
The diagram below shows a typical ELK (Elastic-Logstash-Kibana) stack. Kafka efficiently collects log streams from each instance. ElasticSearch consumes the logs from Kafka and indexes them. Kibana provides a search and visualization UI on top of ElasticSearch.Data streaming in recommendations
E-commerce sites like Amazon use past behaviors and similar users to calculate product recommendations. The diagram below shows how the recommendation system works. Kafka streams the raw clickstream data, Flink processes it, and model training consumes the aggregated data from the data lake. This allows continuous improvement of the relevance of recommendations for each user.System monitoring and alerting
Similar to the log analysis system, we need to collect system metrics for monitoring and troubleshooting. The difference is that metrics are structured data while logs are unstructured text. Metrics data is sent to Kafka and aggregated in Flink. The aggregated data is consumed by a real-time monitoring dashboard and alerting system (for example, PagerDuty).CDC (Change data capture)
Change Data Capture (CDC) streams database changes to other systems for replication or cache/index updates. For example, in the diagram below, the transaction log is sent to Kafka and ingested by ElasticSearch, Redis, and secondary databases.System migration
Upgrading legacy services is challenging - old languages, complex logic, and lack of tests. We can mitigate the risk by leveraging a messaging middleware. In the diagram below, to upgrade the order service in the diagram below, we update the legacy order service to consume input from Kafka and write the result to ORDER topic. The new order service consumes the same input and writes the result to ORDERNEW topic. A reconciliation service compares ORDER and ORDERNEW. If they are identical, the new service passes testing.
Over to you: Do you have any other Kafka use cases to share?
How is data transmitted between applications?
The diagram below shows how a server sends data to another server.
Assume a chat application running in the user space sends out a chat message. The message is sent to the send buffer in the kernel space. The data then goes through the network stack and is wrapped with a TCP header, an IP header, and a MAC header. The data also goes through qdisc (Queueing Disciplines) for flow control. Then the data is sent to the NIC (Network Interface Card) via a ring buffer.
The data is sent to the internet via NIC. After many hops among routers and switches, the data arrives at the NIC of the receiving server.
The NIC of the receiving server puts the data in the ring buffer and sends a hard interrupt to the CPU. The CPU sends a soft interrupt so that ksoftirqd receives data from the ring buffer. Then the data is unwrapped through the data link layer, network layer and transport layer. Eventually, the data (chat message) is copied to the user space and reaches the chat application on the receiving side.
Over to you: What happens when the ring buffer is full? Will it lose packets?An Unusual Request: Combating International Book Piracy on Amazon
As many of you know, I publish my books on Amazon.
It is a great platform to do so. Amazon is where I direct people to find and buy my books. Unfortunately, there is an increasingly problematic piracy issue on the site for my books internationally, especially in India, which I am no longer able to solve by myself. The provided links direct to Amazon India, and ALL the books sold through those links are pirated.
More and more customers are getting low-quality, pirated books shipped to them. The smaller problem is that pirates get paid, and not me. The larger problem is that people get books that are unusable and unacceptable in quality, and leave 1-start reviews.
If you work at Amazon, can you please reply to this email, and help escalate this issue? I would like to keep promoting Amazon as a trusted source to purchase my books. But this issue needs to be resolved, and I'd need help from within the company. Thanks a lot in advance!
Latest articles
Here are the latest articles you may have missed:
To receive all the full articles and support ByteByteGo, consider subscribing:
© 2023 ByteByteGo
548 Market Street PMB 72296, San Francisco, CA 94104
Unsubscribe
by "ByteByteGo" <bytebytego@substack.com> - 11:36 - 9 Sep 2023
Platform | Usecase
