Гипервизоры и уровни виртуализации

Гипервизор – является центральным элементом в виртуализации, выступая как управляющий слой между физическим оборудованием и виртуальными машинами.

С помощью абстрагирование от физических ресурсов, гипервизор обеспечивая гибкое совместное использование и распределение ресурсов. Гипервизоры бывают двух типов:

• Тип 1 (bare-metal): Устанавливаются непосредственно на физический сервер и предоставляют виртуализацию на уровне аппаратного обеспечения.
• Тип 2 (hosted): Устанавливаются на операционную систему хоста и предоставляют виртуализацию на уровне операционной системы.

Концепции виртуализации: many-to-one и one-to-many
Ваше описание двух основных сценариев использования ресурсов в контексте виртуализации и облачных технологий также верно:

• Many-to-one: Этот сценарий описывает ситуацию, когда множество виртуальных машин или контейнеров делят ресурсы одного физического сервера. Это позволяет эффективно использовать доступные ресурсы, увеличивая плотность размещения и снижая затраты.

• One-to-many: В этом случае одно приложение может динамически использовать ресурсы из большого пула, что особенно актуально для приложений, требующих масштабирования в зависимости от нагрузки. Облачные платформы и технологии оркестрации контейнеров, такие как Kubernetes, предоставляют инструменты для управления этим процессом.

Важность абстракции

Абстракция — ключевой элемент виртуализации и облачных технологий, позволяющий отделить прикладной уровень от физического оборудования. Это обеспечивает гибкость, масштабируемость и упрощает управление инфраструктурой. Абстракция позволяет пользователям и приложениям взаимодействовать с виртуальными ресурсами так, как если бы они были физическими, скрывая сложность управления физическими ресурсами.

Понимание виртуализации и облачных технологий через концепции “many-to-one” и “one-to-many” и важность гипервизора как уровня абстракции поможет вам углубить понимание этих технологий на концептуальном уровне.

Практические примеры

Запуск нескольких операционных систем (многие к одному):
Сценарий: вам необходимо одновременно протестировать программное обеспечение в Windows, Linux и macOS.
Решение. Используйте гипервизор, например VMware или VirtualBox, для запуска нескольких виртуальных машин на одной физической машине. Каждая виртуальная машина может иметь другую операционную систему, использующую общие аппаратные ресурсы.

1. Масштабирование приложения (один ко многим):

Сценарий: Ваше веб-приложение испытывает резкий рост трафика.
Решение. Используйте Docker для контейнеризации вашего приложения и Kubernetes для управления контейнерами и их масштабирования на нескольких физических серверах. Kubernetes может динамически распределять ресурсы ЦП, памяти и сети, чтобы обеспечить хорошую работу приложения под нагрузкой.

Пример: развертывание простого приложения с помощью Kubernetes
Давайте рассмотрим процесс развертывания простого веб-приложения с использованием Docker и Kubernetes, продемонстрировав сценарий распределения ресурсов «один ко многим».

Шаг 1. Контейнеризация приложения
Сначала создайте простое веб-приложение Node.js и докерифицируйте его.

app.js:

const express = require('express');
const app = express();
app.get('/', (req, res) => {
  res.send('Hello, World!');
});
const PORT = process.env.PORT || 3000;
app.listen(PORT, () => {
  console.log(`Server is running on port ${PORT}`);
});

Dockerfile:

FROM node:14
WORKDIR /app
COPY package*.json ./
RUN npm install
COPY . .
EXPOSE 3000
CMD ["node", "app.js"]

Создайте образ Docker:

docker build -t my-node-app .

Шаг 2. Отправьте образ Docker в реестр.
Отправьте образ Docker в реестр контейнеров, например Docker Hub, или в частный реестр:

docker tag my-node-app your-dockerhub-username/my-node-app
docker push your-dockerhub-username/my-node-app

Шаг 3. Создайте развертывание и сервис Kubernetes
Создайте манифест развертывания и службы Kubernetes для вашего приложения.

deployment.yaml:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: my-node-app
spec:
  replicas: 3
  selector:
    matchLabels:
      app: my-node-app
  template:
    metadata:
      labels:
        app: my-node-app
    spec:
      containers:
      - name: my-node-app
        image: your-dockerhub-username/my-node-app
        ports:
        - containerPort: 3000

service.yaml:

apiVersion: v1
kind: Service
metadata:
  name: my-node-app
spec:
  selector:
    app: my-node-app
  ports:
    - protocol: TCP
      port: 80
      targetPort: 3000
  type: LoadBalancer

Шаг 4. Развертывание в Kubernetes
Примените манифесты Kubernetes для развертывания приложения:

kubectl apply -f deployment.yaml
kubectl apply -f service.yaml

Kubernetes теперь будет управлять вашим приложением, распределяя нагрузку между несколькими узлами и масштабируя по мере необходимости.

Заключение

Понимание облачных технологий и виртуализации на концептуальном уровне помогает прояснить их сложности. Визуализируя эти концепции как совместное использование ресурсов (многие к одному) и распределение ресурсов (один ко многим), мы можем лучше понять, как работают виртуальные машины и контейнеры, а также их соответствующие преимущества. Эта концептуальная ясность не только помогает понять технологию, но и эффективно применять ее в реальных сценариях.

The post Облачные технологии и виртуализация — ключевые концепции. appeared first on ProdSens.live.

If you’ve been following Discord’s Building Your First Activity tutorial, you may have encountered an issue where the Embedded App SDK gets stuck on the READY event. This issue can be caused by a few different things, but the most common cause is Vite’s hot module reloading (HMR) feature.

We’ve seen this issue pop up in a few different places, so we wanted to provide a quick guide on how to fix it. This post will explain it in more detail and suggest ways around it.

TLDR; Quick Fix

If you’re in a hurry, the quickest way to fix this without disabling HMR is to make sure you’re creating only one instance of the DiscordSDK close to the root of your app. Check out this template to see how it’s done.

You can also create a new Discord Activity project with the fix already included by running this in your terminal:

npx create-robo  -k activity

What’s Happening?

When you’re developing a Discord Activity with Vite, the hot module reloading feature can cause the SDK to be re-created multiple times. This is because Vite reloads the module when you make changes, which can cause the DiscordSDK to be reinitialized.

Unfortunately, discordSdk.ready() gets stuck on the READY event because Discord sends the payload once per instance. That means your activity will be left waiting for the READY event to fire, but it never will until you restart it.

Starting Fresh with a Fix

To fix this issue, you need to ensure that you’re only creating one instance of the DiscordSDK close to the root of your app. This way, the SDK won’t be re-initialized every time you make changes, thus you won’t need to await discordSdk.ready() more than once.

The easiest and fastest way to prevent this is to make sure you start from a template that already has this fix in place. You can clone this template or create a new project with the fix included by running:

npx create-robo  -k activity

This will create a new Discord Activity project with the fix already in place, so you won’t have to worry about this issue. Check out our guide for more info!

➞ Guide: Creating a Discord Activity in seconds with Robo.js

Fixing Your Existing Project

If you’re already working on a project and don’t want to start over, you can fix this issue by ensuring that you’re only creating one instance of the DiscordSDK close to the root of your app. For example, in a React app, you can use a Context to manage the SDK instance.

Here’s a simplified example:

import { DiscordSDK, DiscordSDKMock } from '@discord/embedded-app-sdk'
import { createContext, useContext } from 'react'

export const discordSdk = new DiscordSDK(import.meta.env.VITE_DISCORD_CLIENT_ID)

const DiscordContext = createContext({ discordSdk })

export function DiscordContextProvider(props) {
    const { children } = props
    const setupResult = useDiscordSdkSetup()

    return <DiscordContext.Provider value={setupResult}>{children}DiscordContext.Provider>
}

export function useDiscordSdk() {
    return useContext(DiscordContext)
}

function useDiscordSdkSetup() {
    // Your setup logic here (e.g., discordSdk.ready() in a useEffect)

    return { discordSdk }
}

Then apply the DiscordContextProvider at the root of your app:

import { DiscordContextProvider } from '../hooks/useDiscordSdk'
import { Activity } from './Activity'
import './App.css'

export default function App() {
    return (
        <DiscordContextProvider>
            <Activity />
        DiscordContextProvider>
    )
}

This way, Vite won’t re-initialize the DiscordSDK every time you make changes, and you won’t get stuck on the READY event. It works like so:

1. The DiscordSDK is created once and passed down through the context.
2. The SDK is initialized in the context provider.
3. Components can access the SDK through the custom useDiscordSdk hook.
4. The same SDK instance will be re-used, as long as you don’t edit the useDiscordSdk file.

Conclusion

React’s Context is a great way to manage global state in your app, and it’s perfect for managing the DiscordSDK instance. By ensuring that you only create one instance of the SDK close to the root of your app, you can prevent the READY event from getting stuck and avoid the need to restart your activity every time you make changes.

We highly recommend using create-robo to create a Discord Activity project. These templates are designed to help you get started quickly and avoid common issues like this one. What’s more, they come with a lot of other features and tools to help you build your activity faster and more efficiently, such as Multiplayer Support, TypeScript, Easy Hosting, Streamlined Tunnels, Built-in Database, Plugin System, and so much more!

Don’t forget to join our Discord server to chat with other developers, ask questions, and share your projects. We’re here to help you build amazing apps with Robo.js!

➞ Community: Join our Discord Server

Our very own Robo, Sage, is there to answer any questions about Robo.js, Discord.js, and more

The post How to Fix Stuck Ready Event in Discord Embedded App SDK appeared first on ProdSens.live.

The original post was published on my blog on October 4, 2023.

Hello!

Earlier this week, I wrapped up the “Q3 2023 – Jul-Sep” iteration and cut a new 1.0.0-alpha.3 release of Secutils.dev. In this post, I would like to quickly walk you through the major changes since 1.0.0-alpha.2: notifications, more powerful web page resource tracker, sharing capabilities and more. Let’s dive in!

Scheduled resources checks

If you’ve read my previous posts or tried Secutils.dev web page resources tracker functionality, you might recall that users were required to manually trigger resource checks. With this release, you have an option to schedule automatic resources checks to be performed hourly, daily, weekly, or monthly! When you configure the web page resource tracker, you define how many resource revisions Secutils.dev should store so that you can view the diff between two consecutive revisions. Once the limit is reached, the next revision will displace the oldest one.

Email notifications for changed resources

Since previously you were supposed to manually trigger immediate resource checks, it wouldn’t make much sense to send you any additional notifications about detected changes. You’ll be presented with the check result in the UI as soon as the check is complete without losing context. However, the automatic scheduled resource checks change the control flow, where Secutils.dev should perform the check regularly and notify you if it detects any changes. In the latest release, you can opt in to email notifications, and Secutils.dev will email you if it detects any changes in resources.

Custom resources filtering and mapping

Modern web pages can contain numerous resources, and tracking changes for all of them may not be always necessary. Additionally, certain resources, like those injected by web page analytics solutions, can change with every page load, potentially leading to excessive notifications. In such cases, you’ll likely want to filter out irrelevant resources or focus on specific ones.

In more advanced scenarios, you might be interested in only a portion of a web page resource. For instance, there could be scripts bundling multiple third-party libraries, with changes in some libraries being more important than others. It would be convenient to have the ability to trim or “map” these resources into more meaningful resources.

I explored various approaches to address these use cases in the simplest way possible, but there were always complex edge cases that required a change in direction. However, considering that the primary audience for Secutils.dev is software developers, I decided that introducing some complexity could offer much greater flexibility.

As mentioned in this post, I use Playwright (with Chromium) to extract web page resources. While this choice adds complexity to implementation, security, and deployment, it grants quite a bit of flexibility. With Playwright, I can access, intercept, or modify virtually everything on the tracked web page. Notably, Playwright allows me to inject custom JavaScript scripts into a web page. Rather than inventing my own syntax/parser for custom user resource filters and mapping rules, I can provide users with the full power of modern JavaScript executed within the latest available browser. The only constraint is that users must adhere to the input and output interfaces expected by Secutils.dev.

The potential applications of this approach are vast. I’m already planning to extend it to cover more utilities and use cases, such as tracking changes in page content, not just resources. Imagine a change tracker for virtually anything on the web!

Sharing & collaboration

In today’s world, it’s challenging to envision a software engineer or security researcher working entirely in isolation. As software systems grow in size and complexity, collaboration becomes essential. That’s one of the reasons why collaboration software is on the rise, with built-in collaboration features becoming increasingly common.

While it may be too early to implement full-fledged two-way collaboration functionality in Secutils.dev, I recognize that the absence of such features could limit the tool’s adoption. Therefore, I’m planning to gradually introduce collaboration-related features in each iteration, starting with the “one-way” sharing functionality released in 1.0.0-alpha.3 release. With this release, you can share created content security policies with anyone on the internet, even if they don’t have a Secutils.dev account.

In the future, I intend to expand this sharing functionality to include digital certificate templates and tracked web page resources.

Other enhancements and bug fixes

In addition to the major features mentioned above, this release also includes several smaller enhancements. These include extending the digital certificate editor to allow users to configure private key size (for RSA and DSA) and elliptic curve name (for ECDSA).

As previously mentioned, while the resource tracker functionality has become more powerful, it also comes with increased security risks. Therefore, I’ve made security enhancements for Docker images for all Secutils.dev components, and the Web Scraper component itself. I’ve covered this in more detail in my Running web scraping service securely post.

You can find the full change log here: changelog#1.0.0-alpha.3

In the next few days, I’ll be prioritizing work for the upcoming “Q4 2023 – Oct-Dec” iteration. In my next post, I’ll provide more details on what I’ll be focusing on during this period.

That wraps up today’s post, thanks for taking the time to read it!

ASK: If you found this post helpful or interesting, please consider showing your support by starring secutils-dev/secutils GitHub repository.

Also, feel free to follow me on Twitter, Mastodon, or LinkedIn.

Thank you for being a part of the community!

The post Announcing 1.0.0-alpha.3 release: more powerful resource tracking, notifications and content sharing appeared first on ProdSens.live.

This article is based on Shira’s standout interview with Mark Assini on (fittingly enough!) the Product Marketing Life podcast. Check it out here.

In recent years, major platforms like Spotify and Amazon have invested heavily in podcasting. At the same time, many B2B and B2C brands have launched branded podcasts. So what value do they provide compared to traditional ads?

Well, audio’s immersive intimacy makes podcasts the perfect medium for branded storytelling. Podcasts allow brands to connect with audiences in a deeper way than conventional ads. They should be part of your integrated marketing mix alongside video, social, and email campaigns. 

I’m Shira Atkins, the CRO and Co-founder of Wonder Media Network, an audio-first media company, and I’m here to share my advice on podcasting for brands. I’ll cover topics including:

• How to get started with a branded podcast
• Measuring success
• The advantages of building a podcast network
• How to cut through the noise
• Options beyond launching a podcast
• Tailoring your approach for B2B vs B2C

Let’s jump in.

Key considerations for launching a podcast

Launching a company podcast often seems simpler than it really is. 

To do it right, start by defining your budget and ideal target audience. Conduct market research into what podcasts your audience already listens to. This will help determine the best format, release schedule, hosts, and general approach for your show. For example, decide whether you want interview-based or narrative-style episodes, seasonal or always-on cadence, and so on.

It’s also vital to set realistic expectations upfront. An ongoing weekly interview show requires a big time investment to continually generate fresh content. Be honest about the level of resources you’ll need to produce the podcast you want to create.

The key is understanding your goals, audience, and competitive landscape before you even start recording. That insight will allow you to launch a podcast that truly connects with listeners rather than just ticking a box for your company.

How to measure the performance of your podcast

There are several ways to gauge the success of a branded podcast. Of course, you can look at standard metrics like downloads and impressions – just be sure to tie these back to your sales funnel to see if the podcast helps drive conversions.

Here are some other clever approaches I’ve seen brands use:

• Invite prospective clients on as podcast guests to build relationships. The Trade Desk and Greenhouse HR software do this brilliantly.
• Use the podcast to test storylines and dabble in entertainment. For instance, let’s say you want to put a docu-series together, but first you want to see how the narrative is received – podcasting could be the way to go.
• Align with influencers to reach their audiences. You could even have an influencer host a podcast for you.

Podcasts also provide long-term brand awareness gains that are harder to quantify directly. Ultimately a quality show brings you closer to your audience and helps you stand out from competitors.

The power of podcast networks

Recently, some major B2B brands have launched their own podcast networks rather than a single branded show. They partner with existing creators or acquire podcast content to bring under their umbrella.

What’s the strategic play here? It depends on the specific company and goals, but here are some key advantages:

• Leveraging influencers’ followings: For instance, not so long ago, LinkedIn launched its own podcast network as a way to tap into popular platform users with engaged followings and bring their shows in-house – essentially making them foot soldiers for LinkedIn. 
• Controlling distribution: Owning multiple shows lets LinkedIn feature them all natively on their platform and control the listening experience.
• Showcasing thought leadership: For instance, Klue and HubSpot partner with prominent voices in competitive intelligence and marketing to position themselves as go-to thought leaders and boost their own legitimacy.
• Delivering an in-demand service: Part of what HubSpot promises its users is guidance on how to carry out various marketing tasks. One way it does that is through podcasts.
• Building loyalty: If you’re putting out high-quality content at a regular cadence, that’s a great way to ensure users keep coming back for more. resources to users. 
• Facilitating cross-promotion: A network allows easy cross-collaboration and promotion between shows under one brand umbrella.
• Curating quality content: Finding the best possible podcasts for your niche and then bringing them in-house means users don’t have to trawl Apple Podcasts and hope something good comes up. Instead, you’re doing that work for them. It’s good for the podcasts that are now part of your network, it’s good for your potential clients, and it’s good for you. 

How to make your podcast stand out

It’s harder than ever for podcasts to find an audience amid the explosion of content. Platforms prioritize their own shows, and it’s easy to get lost in the mix. So how can you make sure your branded podcast stands out?

Having a talented, charismatic host is number one. They need to feel approachable and leverage their own existing following. interview people at random.

Also, resist calling it “The [Company] Podcast” or over-focusing on your brand. Hyper-targeted niche topics tend to grab more attention. Find an underserved “job to be done” that your show can uniquely fill, rather than interviewing people at random.

And, of course, providing real entertainment value is key. Useful, buzz-worthy narratives will interest people more than yet another egocentric sales pitch. For example, Pushkin’s branded show “The Legacy” tells a fascinating decades-long story about a track coach – and just happens to feature the sponsor’s CEO occasionally. It works brilliantly because the content stands strongly on its own.

Finally, it may sound silly, but high-quality cover art also matters. Podcasts may seem like an audio-only medium, but people still discover them through visual platforms. A boring stock logo could cause someone to scroll right by. 

In every other aspect of marketing, we sweat the tiniest details of the color palette, the composition, and more. We have to do the same for our cover art. This visual identity can set your show apart subliminally. It’s all part of a cohesive experience.

Podcast partnerships beyond branded shows

If creating your own branded podcast seems too daunting, don’t worry – there are plenty of other ways to engage audiences through podcasting.

The best approach is to strategically partner with relevant shows that reach your target customers. Start by conducting market research to identify which podcasts your audience already listens to. If you’re in B2B, get hyper-targeted. Say you’re in healthcare. Don’t just look at general science or medical podcasts – survey doctors to find which specific shows they tune into.

Once you identify the niche podcasts your customers love, reach out to those show producers about opportunities for sponsorship, partnership, or content integration. Creative options could include:

• 30 or 60-second mid-roll audio ads
• Fully sponsored bonus episodes focused on topics that align with your brand
• Sponsoring limited series within established shows
• Bringing your executives into segments as expert guests
• Participating in live events hosted by the podcast

Podcasters also often have engaged followings outside their audio content, like newsletters and social media groups. Explore promotional collaborations through these channels too.

The key is seamless alignment with podcasts where your positioning makes sense. Don’t force it. Hyper-targeted partnerships can actually be more powerful than trying to launch your own isolated show. Independent creators are generally more nimble and authentic. Plus, supporting indie podcasters is great for the podcasting space as a whole.

Optimizing for B2B vs. B2C audiences

Your podcasting strategy is going to differ depending on whether you’re a B2B or B2C brand.

For B2C, podcast ads can directly drive conversions through concise calls-to-action, discounts, and trackable referral codes. It’s basically just a question of buying ads on shows that your audience listens to.

But it’s best to avoid a pure sales pitch approach with B2B podcasting. Instead, highlight thought leadership, company culture, executive perspectives, and customer stories. They help humanize your brand and build authority over time.

Pfizer takes this branded content approach with their show “Breakthroughs,” which profiles scientists and researchers making strides against disease and tells stories about how they’ve changed patients’ lives. They’re not trying to acquire customers for specific drugs; they’re simply reinforcing their position on the cutting edge of medicine and raising their profile as an employer.

That said, some direct response strategies can work for B2B too. As I mentioned, The Trade Desk invites prospects and clients on as guests to talk about their experience in digital media – that’s a great approach as long as you focus your conversations on their challenges and insights, rather than pushing your product.

A final note on launching your podcast

So there you have it – a guide to making podcasts work for your brand, whether you opt to launch your own show or build strategic partnerships. 

The podcast realm may seem saturated, but if you do it right, you can still grab your audience’s attention and build a connection. At its core, it’s all about crafting stories people genuinely want to hear. Be creative, be bold, and most importantly, be human, and your podcast will pay dividends.

The post Launching a podcast: The key to boosting brand awareness appeared first on ProdSens.live.