Archives
-
Premium PC Components for Seamless Performance
Dear Info,
At Shenzhen Tianfeng International Technology Co., Ltd., we specialize in high-efficiency PC power supplies, durable motherboards, and cutting-edge GPUs designed to meet global OEM/ODM demands. Our products are engineered with advanced thermal management and energy-saving technologies, ensuring reliability under heavy workloads.
Key highlights:
Certified Quality: CE, FCC, and RoHS compliance.
Custom Solutions: Tailored configurations for bulk orders.
Fast Lead Times: Streamlined logistics for timely delivery.
Let’s power innovation together.
Best regards,
Whisper
Mobile:+86 0 18816471391
Whatsapp:+86 18816471391Shenzhen Tianfeng International Technology Co., Ltd.
Website:https://www.skywindintl.com/
802, Fucheng Digital Innovation Park, No. 15, Shijing Road, Fumin Community, Fucheng Street, Shenzhen, Guangdong, China
by "andy" <andy@pcgearshub.com> - 05:40 - 15 May 2025 -
EU Carbon Tax Alert! Optimize [Industry] Waste Treatment Before 2025 Deadline
Dear Info ,
Hello! I am [Marketing center director] from [JIANGSU WORLD TOP THERMAL SCIENCE TECHNOLOGY CO.,LTD]. We specialize in providing advanced environmental technologies and engineering services to global manufacturing industries, helping companies meet environmental requirements while improving production efficiency.
We understand that your company is facing increasing environmental pressures in [Lithium battery manufacturing, pulp manufacturing, seawater desalination for salt production, landfill sites, utilization and disposal of waste incineration fly ash, coal chemical industry, raw material drug production/large-scale pharmaceutical industry, industrial wastewater water treatment, metallurgical industry, utilization of hazardous waste production enterprises]. Our [MVR evaporator, multi effect evaporator, mother liquor dryer, disc dryer, tire rubber cracking device, vibrating fluidized bed, forced circulation heater, condensate preheater, OSLO crystallizer, FC crystallizer, condenser, condensate water tank, thickener, mother liquor tank, fresh steam condensate water tank, scraper film evaporator, plate heat exchanger] can effectively reduce emissions, enhance resource utilization, and help your company achieve lower production costs and higher efficiency.
If you are interested in our environmental technologies, we would be happy to provide you with a detailed solution and discuss the best implementation plan for your company.
Looking forward to hearing from you!
Best regards,
SubscribeUnsubscribe
by "Annual Kiba" <kibaannual@gmail.com> - 02:22 - 15 May 2025 -
[LAST CHANCE] Virtual ticket for SLC Live Event
I hope you’ve been getting my emails about the SLC event…
When: Friday, May 16th 3pm ET
Where: Anywhere with Wi-Fi
You’ll want to be there because even though this is a smaller “Underground Utah Funnel Hacker Event”, Russell is using it to make a MASSIVE announcement!
If I were you, I’d move any meetings you have on your calendar, and set aside a few hours to attend this. It will be solid gold! AND… Russell’s announcing something big.
JOIN US VIRTUALLY FOR MY SALT LAKE CITY HANGOUT EVENT HERE >>
Hope to see you on the Zoom!
(And BTW… If you’re anywhere close to Utah, you should totally make the trip and hang out!)
© Etison LLC
By reading this, you agree to all of the following: You understand this to be an expression of opinions and not professional advice. You are solely responsible for the use of any content and hold Etison LLC and all members and affiliates harmless in any event or claim.
If you purchase anything through a link in this email, you should assume that we have an affiliate relationship with the company providing the product or service that you purchase, and that we will be paid in some way. We recommend that you do your own independent research before purchasing anything.
Copyright © 2018+ Etison LLC. All Rights Reserved.
To make sure you keep getting these emails, please add us to your address book or whitelist us. If you don't want to receive any other emails, click on the unsubscribe link below.
Etison LLC
3443 W Bavaria St
Eagle, ID 83616
United States
by "Todd Dickerson" <noreply@clickfunnelsnotifications.com> - 12:50 - 15 May 2025 -
Engineering Trade-offs: Eventual Consistency in Practice
Engineering Trade-offs: Eventual Consistency in Practice
Forwarded this email? Subscribe here for moreLatest articles
If you’re not a subscriber, here’s what you missed this month.
Coupling and Cohesion: The Two Principles for Effective Architecture
OOP Design Patterns and Anti-Patterns: What Works and What Fails
The Art of REST API Design: Idempotency, Pagination, and Security
To receive all the full articles and support ByteByteGo, consider subscribing:
Imagine a ride-sharing app that shows a driver’s location with a few seconds of delay. Now, imagine if the entire app refused to show anything until every backend service agreed on the perfect current location. No movement, no updates, just a spinning wheel.
That’s what would happen if strong consistency were always preferred in a distributed system.
Modern applications (social feeds, marketplaces, logistics platforms) don’t run on a single database or monolithic backend anymore. They run on event-driven, distributed systems. Services publish and react to events. Data flows asynchronously, and components update independently. This decoupling unlocks flexibility, scalability, and resilience. However, it also means consistency is no longer immediate or guaranteed.
This is where eventual consistency becomes important.
Some examples are as follows:
A payment system might mark a transaction as pending until multiple downstream services confirm it.
A feed service might render posts while a background job deduplicates or reorders them later.
A warehouse system might temporarily oversell a product, then issue a correction as inventory updates sync across regions.
These aren’t bugs but trade-offs.
Eventual consistency lets each component do its job independently, then reconcile later. It prioritizes availability and responsiveness over immediate agreement.
This article explores what it means to build with eventual consistency in an event-driven world. It breaks down how to deal with out-of-order events and how to design systems that can handle delays.
What is Eventual Consistency?...
Continue reading this post for free in the Substack app
Like
Comment
Restack
© 2025 ByteByteGo
548 Market Street PMB 72296, San Francisco, CA 94104
Unsubscribe
by "ByteByteGo" <bytebytego@substack.com> - 11:37 - 15 May 2025 -
Your invited! New Relic University - Observability Workshop Series
Hi MD,
Phil Weber, Lead Customer Training Specialist from New Relic, here.
New Relic Certification is now available to help you continue to grow your observability knowledge throughout your career. I am running a series of workshops designed to upskill your observability skills with New Relic, and help you get the skills and knowledge to take your chosen exams. We’ll also dive into some hands-on labs where you’ll work within a sandbox in a live working environment to put your skills to the test.
Dates and topics:
- 20 May: Maximise performance engineering with APM and infrastructure monitoring - Learn powerful APM and infrastructure monitoring techniques to track, spot and visualise overall system health, and explore how to leverage these insights to make informed, data-driven decisions.
- 21 May: Enhance reliability engineering by resolving incidents faster with New Relic alerts and AI - Explore the fundamentals of New Relic alerts and artificial intelligence, focusing on equipping you with the skills to proactively implement, configure, and optimise for system reliability.
- 22 May: Boost performance & reliability engineering with dashboard visualisation techniques - Learn dashboard design strategies, access best in class templates, discover how to create powerful charts and how to correlate different data sets.
Unable to make it? No sweat. Register here and an on-demand recording of the session will be sent to you after the webinar.
I hope to see you then,
Phil Weber,Lead Customer Training SpecialistNew Relic
This email was sent to info@learn.odoo.com · · View this online
This email was sent to info@learn.odoo.com. If you no longer wish to receive these emails, click on the following link: Unsubscribe
by "Phil Weber, New Relic" <emeamarketing@newrelic.com> - 05:02 - 15 May 2025 -
Can better reading make you a better leader?
On McKinsey Perspectives
by "Only McKinsey Perspectives" <publishing@email.mckinsey.com> - 01:12 - 15 May 2025 -
Emerging Markets Custom Solutions | Inventory Risk Control
Dear Info,
In the face of rapidly shifting demands and inventory overstock risks in emerging markets, we, as an OEM/ODM manufacturer, provide end-to-end solutions to empower your growth:
✓ Flexible Small-Batch Production
- MOQ 1,000 units | 5-day rapid turnaround (mixed SKUs supported)
- Custom packaging/flavors/hardware specs (e.g., humidity-proof versions)
✓ Dynamic Inventory Management
- Pre-sale model + safety stock alerts
- 20% order value for overstock compensation
- Free return with new products on extra discount in terms of the overstock problem
✓ Localized logistic and marketing Support
- Global warehouses to support you with faster delivery
- Customized bundles (e.g. 1/2/3 pod + 3/6/9 cartridges)
Success Story:
In 2024, our Russian partner boosted top-selling SKU ratio from 45% to 82% with data-driven product selection guidance, cutting inventory turnover time by 15 days.
Act Now:
? Claim Your Free Sample (First-order exclusive)
? Download attached our E-catalog and tell us the products of your interests
Looking forward to hearing from you.Best regards,
Yeser
Sales manager
Tel：+86 135 0280 4383
by "Sava Wheatley" <savawheatley@gmail.com> - 05:33 - 14 May 2025 -
Consolidation&Warehouse Services covering all the China port
Dear Info,
Smartrans International Ltd is a freight-forwarding agency based in Shenzhen China, Services as belows.
1)Professional Ocean&AIR freight transportation handing
2)Professional Ocean&AIR freight ground handling agent, Ocean & Airr freight consolidator
3)Worldwide door to door service by SEA or by AIR
4)Common Sea & air freight warehousing and Sea & air freight handling for forwarders
5)Customs clearance Import&Export
6)Cheaking the package at our warehouse
7)Air freight charter from China main cities to worldwide handling
8)FBA shipping
9)Import&Export License documents purchasing
by "Marsha Bendetti" <marshabendetti54@gmail.com> - 03:54 - 14 May 2025 -
[VIRTUAL TICKET] Hang w/ Russell in Utah!
Russell only had room for about 200 people at his SLC event in two days…
BUT… you can still hang with me virtually in Utah if you hurry >>
Even though they couldn’t squeeze a lot of people in the room, Russell was adamant that we have to live stream this because he wants to make a HUGE announcement!
So, you’ll be there, right!?
When: Friday, May 16th 3pm ET
Where: Anywhere with Wi-Fi
Get your ticket and “virtually” join us in Utah!
JOIN US VIRTUALLY FOR THE SALT LAKE CITY HANGOUT EVENT HERE >>
See you there!
© Etison LLC
By reading this, you agree to all of the following: You understand this to be an expression of opinions and not professional advice. You are solely responsible for the use of any content and hold Etison LLC and all members and affiliates harmless in any event or claim.
If you purchase anything through a link in this email, you should assume that we have an affiliate relationship with the company providing the product or service that you purchase, and that we will be paid in some way. We recommend that you do your own independent research before purchasing anything.
Copyright © 2018+ Etison LLC. All Rights Reserved.
To make sure you keep getting these emails, please add us to your address book or whitelist us. If you don't want to receive any other emails, click on the unsubscribe link below.
Etison LLC
3443 W Bavaria St
Eagle, ID 83616
United States
by "Todd Dickerson" <noreply@clickfunnelsnotifications.com> - 01:33 - 14 May 2025 -
Pro Tip: Register NOW to watch live…OR later!
Cross that finish line!
We noticed that you have not registered for SAP Sapphire Virtual yet! We’d hate for you to miss out! Whether you watch live or watch later, register NOW to customize your event experience – and get exclusive access to on‑demand recordings and more.
👉 Visit the website to view the Session Catalog and Event Agenda.
What you can expect:
•
Two live channels plus on‑demand content
•
Keynotes, breaking news, and announcements
•
Customer stories from KIND, Mercedes‑Benz, Standard Chartered Bank, and more
•
Exclusive interviews with SAP customers, executives, and experts
•
Demos with Q&A, road map reviews, and strategy talks
Ready to get the “best-of” SAP Sapphire?
Looking for more information
about our SAP Sapphire event program?
SAP (Legal Disclosure | SAP)
This e-mail may contain trade secrets or privileged, undisclosed, or otherwise confidential information. If you have received this e-mail in error, you are hereby notified that any review, copying, or distribution of it is strictly prohibited. Please inform us immediately and destroy the original transmittal. Thank you for your cooperation.You are receiving this e-mail for one or more of the following reasons: you are an SAP customer, you were an SAP customer, SAP was asked to contact you by one of your colleagues, you expressed interest in one or more of our products or services, or you participated in or expressed interest to participate in a webinar, seminar, or event. SAP Privacy Statement
This e-mail was sent to you on behalf of the SAP Group with which you have a business relationship. If you would like to have more information about your Data Controller(s) please click here to contact webmaster@sap.com.
This promotional e-mail was sent to you by SAP Global Marketing and provides information on SAP's products and services that may be of interest to you. If you would prefer not to receive such e-mails from SAP in the future, please click on the Unsubscribe link.
To ensure you continue to receive SAP related information properly please add sap@mailsap.com to your address book or safe senders list.
by "SAP Sapphire" <sap@mailsap.com> - 11:10 - 14 May 2025 -
Photonics/ Photoelectric/Optical Products from Qualphot Inc.
Dear Sir/Madam,
Qualphot Inc. design & manufacture Photonics/ Photoelectric/Optical Products. Our products have advantages in specifications and quality.
We can support you a competitive price and very fast lead time for most of our products.Our main products:
1. Photodetectors: High Sensitivity/ High Speed/ Low Noise/ APD/ Balanced/ Multi-channel Photodetectors
2. Light Sources/ Lasers: Narrow Linewidth/ Broadband/ Direct-Modulation/ DFB/ Tunable Light Sources
3. Modulators: Electro Optic Intensity/ Electro Optic Phase/ Y Wavguide Modulators
4. Optical Amplifiers: Erbium Doped Fiber (EDFA)/ Ytterbium Doped Fiber (YDFA)/ Semiconductor Optical (SOA) Amplifiers
5. RoF Links: Optical Transmitter Modules, Optical Receiver Modules, Microwavve Optical Transmission Systems
6. Light Polarization and Delay Series: Fiber Optical Delay Devices, Faraday Mirrors, Switches, Attenuators, Fiber Collimators, WDMs, Polarization Controllers
7. Optical Measuring Instruments: Polarization Extinction Ratio Testers, Multichannel Optical Power Meters, Broadband Light Sources, Small Spectrometer, Frequency Sources
8. Optical Passive Devices: Beam Splitters/Couplers, Circulators/ Isolators, ColimatorsPlease refer to our official website to check information of our standard products. Website URL is in my contact info area bellow.
If you do not find the suitable products, we can also customize it for you.
If you have any question, please feel free to contact us. I look forward to hear from you for future cooperation. Thank you!
Best Regards,
--
D.M. He
Qualphot Inc.
RM502, Building 1-1, Tianke Plaza, No. 999 Tiangong Ave.,
Tianfu New Area, Chengdu, 610213, China
Tel: +86-28-6115-4929
E-mail: sales@qualphot.com
Website: www.qualphot.com
by "Ritu Gound" <ritugound508@gmail.com> - 08:23 - 14 May 2025 -
High Pressure 100L - 3000L Hot water tank buffer tank
Hi Info，
We manufacture and export full range of hot water tanks and buffer tanks, products include heat pump pressure water tanks, buffer tanks,boiler water tanks, coil heat exchange water tanks, solar water heater tanks, electric water heater tanks, etc. SolarShine have more than 15 years producing history, our products are certified with CE, ISO9001, TUV and SGS certifications.
If you’re interested in getting more detail or price, kindly let us know soon. Thank you.
Kind regards,
Coco Li/ Export Manager
SHENZHEN SOLARSHINE RENEWABLE ENERGY TECHNOLOGY CO.,LTD.
Rida Industrial Zone, Number 61, Longxin, Pingshan District, Shenzhen, China
http://www.solarshine01.com
WeChat/whatsapp: 0086 1860 6513 895
Tel: +86 0755 - 8411 4281
by "cocoli.heatpump" <cocoli.heatpump@heatpump-chiller.com> - 08:01 - 14 May 2025 -
Transform Traditional Meters into Smart Meters with Our Innovative Solutions
Dear Sir/Madam,
Are you looking to upgrade your traditional mechanical meters to smart meters? HAC Telecom has helped numerous companies quickly transition into the smart meter market, providing them with efficient, reliable solutions tailored to their needs. Our electronic backpack (pulse reader) is specifically designed for retrofitting traditional meters, seamlessly integrating with leading brands such as Sensus, Insa, and NWM.
I’d love to explore how we can help your business make the switch to smart meters. Please feel free to reply to this email, and we can discuss your needs in detail.
Best regards,
Winni
Senior Export Manager
Shenzhen HAC Telecom Technology Co., Ltd.
Phone ：+86 18565749800
Mail： liyy@rf-module-china.com
Website：www.rf-module-china.com
www.haccom.cn
by "Yanik Oleksy" <oleksyyanik@gmail.com> - 07:04 - 14 May 2025 -
Gen Z wants office time. Moms want flexibility. Leaders need to navigate both.
On McKinsey Perspectives
Brought to you by Alex Panas, global leader of industries, & Axel Karlsson, global leader of functional practices and growth platforms
Welcome to the latest edition of Only McKinsey Perspectives. We hope you find our insights useful. Let us know what you think at Alex_Panas@McKinsey.com and Axel_Karlsson@McKinsey.com.
—Alex and Axel
•
Working more flexibly. Are offices packing up to the rafters again? Not exactly. McKinsey’s American Opportunity Survey finds a more nuanced reality: While there has been an uptick in respondents reporting they work fully on-site compared with 2022, the change has been modest, McKinsey Senior Partner Nora Gardner and coauthors share. McKinsey Global Institute research has found that, on average, sweeping changes brought on by the pandemic have largely stabilized: Workers now go into the office about 30% less often than they did prior to the pandemic.
•
Who prefers remote work? If employees’ preferences alone shaped remote-work policies, these policies would likely expand, as more than half of surveyed workers said they favor remote-work models. Those with higher education and income levels were both more likely to favor and secure remote roles compared with other cohorts. In contrast, the youngest workers surveyed (aged 18 to 25) were the least enthusiastic about remote work. Learn how employers and real estate companies can adapt to this new era of work.
—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 Only McKinsey Perspectives newsletter, formerly known as Only McKinsey.
Copyright © 2025 | McKinsey & Company, 3 World Trade Center, 175 Greenwich Street, New York, NY 10007
by "Only McKinsey Perspectives" <publishing@email.mckinsey.com> - 01:44 - 14 May 2025 -
Big Datadog announcements coming at DASH
The future of observability is here.AI is transforming the way we monitor, secure, and build modern applications–and Datadog is at the forefront of this shift.
Join us for the DASH keynote and be the first to hear the latest announcements about new products and features across the Datadog platform. From preventing incidents before they escalate to automating resolution workflows, we’re delivering powerful AI-driven capabilities to help you scale faster and operate more securely.We’ll share how we’re thoughtfully leveraging AI in the tools you use today, and introduce new ones to help you:
- Observe any stack at any scale
- Secure your AI workloads and applications–from code to data to the cloud
- Take action through agentic systems to detect and respond to issues in real time
See you in NYC!
The DASH Team
Visit our email preference center to unsubscribe or update subscription options© 2025 Datadog Inc, 620 8th Ave, 45th Floor, New York, NY 10018This email was sent to info@learn.odoo.com.
by "DASH by Datadog" <dash@datadoghq.com> - 10:30 - 13 May 2025 -
How Slack Supports Billions of Daily Messages
How Slack Supports Billions of Daily Messages
Forwarded this email? Subscribe here for moreGenerate your MCP server with Speakeasy (Sponsored)
Like it or not, your API has a new user: AI agents. Make accessing your API services easy for them with an MCP (Model Context Protocol) server. Speakeasy uses your OpenAPI spec to generate an MCP server with tools for all your API operations to make building agentic workflows easy.
Once you've generated your server, use the Speakeasy platform to develop evals, prompts and custom toolsets to take your AI developer platform to the next level.
Disclaimer: The details in this post have been derived from the articles/videos shared online by the Slack engineering team. All credit for the technical details goes to the Slack Engineering Team. The links to the original articles and videos are present in the references section at the end of the post. We’ve attempted to analyze the details and provide our input about them. If you find any inaccuracies or omissions, please leave a comment, and we will do our best to fix them.
Most people think of Slack as a messaging app. It is technically accurate, but from a systems perspective, it's more like a real-time, multiplayer collaboration platform with millions of concurrent users, thousands of messages per second, and an architecture that evolved under some unusual constraints.
At peak weekday hours, Slack maintains over five million simultaneous WebSocket sessions. That’s not just a metric, but a serious architectural challenge. Each session represents a live, long-running connection, often pushing out typing indicators, presence updates, and messages in milliseconds. Delivering this kind of interactivity on a global scale is hard. Doing it reliably with high performance is even harder.
One interesting trivia is that the team that built Slack was originally building a video game named Glitch: a browser-based MMORPG. While Glitch had a small but passionate audience, it struggled to become financially sustainable. During the development of Glitch, the team created an internal communication tool that would later become Slack. When Glitch shut down, the team recognized the potential of the internal communication tool and began to develop it into a bigger product for business use. The backend for this internal tool became the skeleton of what would become Slack.
This inheritance shaped Slack’s architecture in two key ways:
Separation of concerns: Like game servers manage real-time events separately from game logic, Slack splits its architecture early. One service (the “channel server”) handled real-time message propagation. Another (the “web app”) managed business logic, storage, and user auth.
Push-first mentality: Unlike traditional request-response apps, Glitch pushed updates as the state changed. Slack adopted this model wholesale. WebSockets weren’t an optimization—they were the foundation.
This article explores how Slack’s architecture evolved to meet the demands of a system that makes real-time collaboration possible across organizations of 100,000+ people.
Real-Time Code Reviews Powered by AI (Sponsored)
Slack made real-time collaboration seamless for teams. CodeRabbit brings that same spirit to code reviews. It analyzes every PR using context-aware AI that understands your codebase suggesting changes, catching bugs and even asking questions when something looks off. Perfect for fast-moving teams who want quality code reviews without slowing down. Integrated with GitHub, GitLab and it's like a senior engineer reviewing with you on every commit. Free for Open-source.
Review Smarter with CodeRabbit
Initial Architecture
Slack’s early architecture was a traditional monolithic backend fused with a purpose-built, real-time message delivery system.
The monolith, written in Hacklang, handled the application logic. Hacklang (Facebook’s typed dialect of PHP) offered a pragmatic path: move fast with a familiar scripting language, then gradually tighten things with types. For a product iterating quickly, that balance paid off. Slack’s backend handled everything from file permissions to session management to API endpoints.
But the monolith didn’t touch messages in motion. That job belonged to a real-time message bus: the channel server, written in Java. The channel server pushed updates over long-lived WebSocket connections, broadcast messages to active clients, and arbitrated message order. When two users hit “send” at the same moment, it was the channel server that decided which message came first.
Here’s how the division looked in terms of functionalities:
Web App (Hacklang)
Auth, permissions, and storage
API endpoints and job queuing
Session bootstrapping and metadata lookup
Channel Server (Java)
WebSocket handling
Real-time message fan-out
Typing indicators, presence blips, and ordering guarantees
This split worked well when Slack served small teams and development moved fast. But over time, the costs surfaced:
The monolith grew brittle as testing got harder and deployment risk went up.
The channel server held state, which complicated recovery and scaling.
Dependencies between the two made failures messy. If the web app went down, the channel server couldn’t persist messages, but might still tell users they’d sent them.
The Core Abstraction: Persistent Messaging
Messaging apps live or die by trust. When someone sends a message and sees it appear on screen, they expect it to stay there and to show up for everyone else. If that expectation breaks, the product loses credibility fast. In other words, persistence becomes a foundational feature.
Slack’s design bakes this in from the start. Unlike Internet Relay Chat (IRC), where messages vanish the moment they scroll off-screen, Slack assumes every message matters, even the mundane ones. It doesn’t just aim to display messages in real-time. It aims to record them, index them, and replay them on demand. This shift from ephemeral to durable changes everything.
IRC treats each message like a radio transmission, whereas Slack treats messages like emails. If the user missed something, they can always scroll up, search later, and re-read at a later date. This shift demands a system that guarantees:
Messages don’t disappear
Message order stays consistent
What one user sees, every user sees
Slack delivers that through what looks, at first glance, like a simple contract:
When a message shows up in a channel, everyone should see it.
When a message appears in the UI, it should be in stable storage.
When clients scroll back, they should all see the same history, in the same order.
This is a textbook case of atomic broadcast.
Atomic Broadcast
Atomic broadcast is a classic problem in distributed systems. It's a formal model where multiple nodes (or users) receive the same messages in the same order, and every message comes from someone. It guarantees three core properties:
Validity: If a user sends a message, it eventually gets delivered.
Integrity: No message appears unless it was sent.
Total Order: All users see messages in the same sequence.
Slack implements a real-world approximation of atomic broadcast because it was essential for their functionality. Imagine a team seeing different sequences of edits, or comments that reference messages that “don’t exist” on someone else’s screen.
But here’s the twist: in distributed systems, atomic broadcast is as hard as consensus. And consensus, under real-world failure modes, is provably impossible to guarantee. So Slack, like many production systems, takes the pragmatic path. It relaxes constraints, defers work, and recovers from inconsistency instead of trying to prevent it entirely.
This tension between theoretical impossibility and practical necessity drives many of Slack’s architectural decisions.
Old vs New Send Flows
In real-time apps, low latency is a necessity. When a user hits “send,” the message should appear instantly. Anything slower breaks the illusion of conversation. But making that feel snappy while also guaranteeing that the message is stored, ordered, and replayable? That’s where things get messy.
Slack’s original message send flow prioritized responsiveness. The architecture puts the channel server (the real-time message bus) at the front of the flow. A message went from the client straight to the channel server, which then:
Broadcast it to all connected clients
Sent an acknowledgment back to the sender
Later handed it off to the web app for indexing, persistence, and other deferred work
This gave users lightning-fast feedback. However, it also introduced a dangerous window: the server might crash after confirming the message but before persisting it. To the sender, the message looked “sent.” To everyone else, especially after a recovery, it might be gone.
This flow worked, but it carried risk:
Stateful servers meant complex failover logic and careful coordination.
Deferred persistence meant the UI could technically lie about message delivery.
Retries and recovery had to reconcile what was shown vs. what was saved.
Slack patched around this with persistent buffers and retry loops. But the complexity was stacking up. The system was fast, but fragile.
The Web App Takes the Lead
As Slack matured, and as outages and scale pushed the limits, the team reversed the flow.
In the new send model, the web app comes first:
The client sends the message via HTTP POST to the web app
The web app logs the message to the job queue (persistence, indexing, and parsing all happen here)
Only then does it invoke the channel server to broadcast the message in real-time
This change improves several things:
Crash safety: If anything goes down mid-flow, either the message persists or the client gets a clear failure.
Stateless channel servers: Without needing local buffers or retries, they become easier to scale and maintain.
Latency preserved: Users still see messages immediately, because the real-time broadcast happens fast, even while persistence continues in the background.
And one subtle benefit: the new flow doesn’t require a WebSocket connection to send a message. That’s a big deal for mobile clients responding to notifications, where setting up a full session just to reply was costly.
The old system showed messages fast, but sometimes dropped them. The new one does more work up front, but makes a stronger promise in terms of persistence.
Session Initialization and the Need for Flannel
For small teams, starting a Slack session looks simple. The client requests some data, connects to a WebSocket, and starts chatting. However, at enterprise scale, that “simple” startup becomes a serious architectural choke point.
Originally, Slack used a method called RTM Start (Real-Time Messaging Start). When a client initiated a session, the web app assembled a giant JSON payload: user profiles, channel lists, membership maps, unread message counts, and a WebSocket URL. This was meant to be a keyframe: a complete snapshot of the team’s state, so the client could start cold and stay in sync via real-time deltas.
It worked until teams got big.
For small teams (under 100 users), the payload was lightweight.
For large organizations (10,000+ users), it ballooned to tens of megabytes.
Clients took tens of seconds just to parse the response and build local caches.
If a network partition hit, thousands of clients would reconnect at once, slamming the backend with redundant work.
And it got worse:
Payload size grew quadratically with team size. Every user could join every channel, and the web app calculated all of it.
All this work happened in a single data center, creating global latency for users in Europe, Asia, or South America.
This wasn’t just slow. It was a vector for cascading failure. One bad deploy or dropped connection could take out Slack’s control plane under its load.
Flannel: Cache the Cold Start
To fix this, Slack introduced Flannel, a purpose-built microservice that acts as a stateful, geo-distributed cache for session bootstrapping.
Instead of rebuilding a fresh session snapshot for every client on demand, Flannel does a couple of things differently:
It maintains a pre-warmed in-memory cache of team metadata
Listens to WebSocket events to keep that cache up to date, just like a client would
It serves session boot data locally, from one of many regional replicas
Sits astride the WebSocket connection, terminating it and handling session validation
Here’s what changes in the flow:
A client connects to Flannel and presents its auth token.
Flannel verifies the token (delegating to the web app if needed).
If the cache is warm, it sends a hello response immediately. No need to hit the origin.
This flips the cost model from compute-heavy startup to cache-heavy reuse. While it’s tempting to think that Flannel adds complexity. But Slack found that at scale, complexity that’s predictable and bounded is better than simplicity that breaks under pressure.
Scaling Considerations and Trade-offs
Every system seems to work on the whiteboard. The real test comes when it’s live, overloaded, and something fails. At Slack’s scale, maintaining reliable real-time messaging isn’t just about handling more messages per second. It’s also about absorbing failure without breaking user expectations.
One of the most visible symptoms at scale is message duplication. Sometimes a user sees their message posted twice. It’s not random. It’s a side effect of client retries.
Here’s how it happens:
A mobile client sends a message.
Network flakiness delays the acknowledgment.
The client times out and retries.
Both messages make it through, or one makes it twice, and now the user wonders what just happened.
To survive this, Slack leans on idempotency. Each message includes a client-generated ID or salt. When the server sees the same message ID again, it knows it’s not a new send. This doesn’t eliminate all duplication, especially across devices, but it contains the damage.
On the backend, retries and failures get more serious. A message might:
Reach the channel server but fail to persist
Persist to the job queue but never push
Push to some clients and not others
The system has to detect and recover from all of these without losing messages, breaking order guarantees, and flooding the user with confusing errors.
This is where queueing architecture matters. Slack uses Kafka for durable message queuing and Redis for in-flight, fast-access job data. Kafka acts as the system’s ledger and Redis provides short-term memory.
This separation balances:
Durability vs. speed: Kafka holds the truth; Redis handles the work-in-progress.
Retry logic: Jobs pulled from Kafka can be retried intelligently if processing fails.
Concurrency control: The system avoids processing the same message twice and waiting forever for a stuck job.
Conclusion
Slack’s architecture isn’t simple, and that’s by design. The system embraces complexity in the places where precision matters most: real-time messaging, session consistency, and user trust. These are the end-to-end paths where failure is visible, consequences are immediate, and user perception can shift in a heartbeat.
The architecture reflects a principle that shows up in high-performing systems again and again: push complexity to the edge, keep the core fast and clear. Channel servers, Flannel caches, and job queues each exist to protect a smooth user experience from the messiness of distributed systems, partial failures, and global scale.
At the same time, the parts of the system that don’t need complexity, like storage coordination or REST API responses, stay lean and conventional.
Ultimately, no architecture stands still. Every scaling milestone, every user complaint, every edge case pushes the system to adapt. Slack’s evolution from monolith-plus-bus to globally distributed microservices wasn’t planned in a vacuum. It came from running into real limits, then designing around them.
The lesson isn’t to copy Slack’s architecture. It’s to respect the trade-offs it reveals:
Optimize for latency, but tolerate slowness in the right places.
Build around failure, not away from it.
Embrace complexity where correctness pays for itself, and fight to simplify the rest.
References:
SPONSOR US
Get your product in front of more than 1,000,000 tech professionals.
Our newsletter puts your products and services directly in front of an audience that matters - hundreds of thousands of engineering leaders and senior engineers - who have influence over significant tech decisions and big purchases.
Space Fills Up Fast - Reserve Today
Ad spots typically sell out about 4 weeks in advance. To ensure your ad reaches this influential audience, reserve your space now by emailing sponsorship@bytebytego.com.
© 2025 ByteByteGo
548 Market Street PMB 72296, San Francisco, CA 94104
Unsubscribe
by "ByteByteGo" <bytebytego@substack.com> - 11:38 - 13 May 2025 -
IDC’s first European observability report
New Relic
.png)
The first-ever IDC MarketScape: European Observability Market 2025 Vendor Assessment names New Relic a leader in strategic vision and capabilities.
The report highlights how New Relic helps companies achieve an 88% reduction in unplanned downtime, along with these benefits:
- Programmable platform to build visualisations, click paths, and more tailored to your business needs.
- More KPIs reached—such as lower MTTR, higher customer satisfaction, and less churn—using observability insights.
- Advanced AI including the ability to monitor LLMs and agentically automate operations.
- Strong European presence and understanding of local needs.
Download the report to see how our customisable platform helps leading European organisations break down silos, streamline operations, and deliver exceptional digital experiences.
Read Our Profile
by "New Relic" <emeamarketing@newrelic.com> - 05:03 - 13 May 2025 -
Why the world needs balance and trust for the economy to thrive
On McKinsey Perspectives
Brought to you by Alex Panas, global leader of industries, & Axel Karlsson, global leader of functional practices and growth platforms
Welcome to the latest edition of Only McKinsey Perspectives. We hope you find our insights useful. Let us know what you think at Alex_Panas@McKinsey.com and Axel_Karlsson@McKinsey.com.
—Alex and Axel
•
Rebuilding confidence. Since the US tariff announcements in early April, many analysts have raised their estimates of the probability of a worldwide recession. For leaders, the question now isn’t whether the current path of escalating tariffs is the right one; rather, it is whether the path they choose next will foster trust among countries and within societies. For many, the answer may be for global and local economies to find new balance: a US that produces more of what it consumes, a China that consumes more of what it produces, a more competitive Europe, and a “global south” that finds its path to prosperity, McKinsey Global Institute Chair Sven Smit 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 Only McKinsey Perspectives newsletter, formerly known as Only McKinsey.
Copyright © 2025 | McKinsey & Company, 3 World Trade Center, 175 Greenwich Street, New York, NY 10007
by "Only McKinsey Perspectives" <publishing@email.mckinsey.com> - 01:13 - 13 May 2025 -
Revolutionary innovations propelling growth
Share this email
New from McKinsey Quarterly
•
•
Access the interactive with a free digital Quarterly membership, which in this commemorative year includes special issues, bonus digital features, and classic articles previously available only in print. Plus, get past digital Quarterly issues and downloads of nearly 100 important reports in The McKinsey Insights Store.
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 © 2025 | McKinsey & Company, 3 World Trade Center, 175 Greenwich Street, New York, NY 10007
by "McKinsey Quarterly" <publishing@email.mckinsey.com> - 02:56 - 12 May 2025
