Technology Co., Ltd. All rights reserved. 云原生应用可观测性实践 向阳 @ 云杉网络 2021-12-08 simplify the growing complexity © 2021, YUNSHAN Networks Technology Co., Ltd. All rights reserved. 可观测性 - What & Why 云原生社区可观察性SIG-定义 demo! 目录 simplify the growing complexity © 2021, YUNSHAN Networks Technology Co., Ltd. All rights reserved. 可观测性的成熟度模型 1.0 基础支柱 2.0 ? 3.0 ? simplify the growing complexity © 2021, YUNSHAN Networks Technology growing complexity © 2021, YUNSHAN Networks Technology Co., Ltd. All rights reserved. 开箱即用的解决方案 Metrics, tracing, and logging 2017.02.21 Peter Bourgon simplify the growing complexity © 2021,

统一数据平台的落地思路及案例 构建统一的云原生应用可观测性数据平台 看云网更清晰 Simplify the growing complexity. 统一的可观测性数据平台 telegraf 看云网更清晰 Simplify the growing complexity. 挑战：数据孤岛、资源开销 数据 孤岛 资源消耗 telegraf 1. 可观测性数据平台的挑战 2. 解决数据孤岛：AutoTagging 统一数据平台的落地思路及案例 构建统一的云原生应用可观测性数据平台 看云网更清晰 Simplify the growing complexity. OpenTelemetry的方法 统一的上下文 以追踪为核心 看云网更清晰 Simplify the growing complexity. OpenTelemetry的方法 Tag, Exemplars (TraceID, SpanID) Tag & Tag 看云网更清晰 Simplify the growing complexity. 数据打通并不简单 ① Trace与「非Request scope」的Metrics 例如：响应Request A的实例在一段时间内做了多少次GC？ ① 看云网更清晰 Simplify the growing complexity. 数据打通并不简单 ② 应用、系统、网络的Metrics之间 例

September 6, 2023 50 / 57 The Curse of Dimensionality Handling complexity Involve many variables, how can we handle this complexity without get- ting into trouble. Optimization and Integration Usually Overview September 6, 2023 51 / 57 How to Handle Complexity Properly dealing with complexity is a crucial issue for machine learning. Limiting complexity is one approach Use a model that is complex enough if can find out how to reduce the large number of variables to a small number. Averaging over complexity is the Bayesian approach. Use as complex a model might be needed, but don’t choose a single parameter

to the linear computation complexity of RNNs. However, attention is still faster in wall clock time because it processes entire sequences together. The quadratic complexity of attention is addressed through transformer variants with efficient self-attention mechanisms. These ideas tackle the quadratic complexity at various levels. The simplest idea is to chunk the input sequence of length n into blocks of the input sequence, thus compressing it in the process. Sparse attention attempts to reduce the complexity by picking a subset of parameters for attention computation. This subset can also be learned during

cost) ○ Enterprise team structure gap (Workspace, Tenants, etc) ○ UI&UX Background ● Leads to complexity and lack of operational agility ● You can't be Cloud Native at scale without a modern application- Tetrate Academy Warp up • We built products on top of the upstream Istio. • We aim to solve the complexity of Istio and build a zero-trust network for application connectivity. • We are committed to maintaining

manifest transferring blobs pushing manifest finished has tags No Yes The Complexity of Replication Job •The complexity adds up in these aspects: • Monitoring (logging) • Error handling • Arbitrary

style 3 Agenda Day2 Operation Challenge of Kubernetes Day 2 Operation Tame operational complexity Observability Observability Demo 3 Day2 運營 定義與說明 Kubernetes Day2 運營的挑戰 馴服運營 複雜性 可觀測性 K8S Operation Survey @2021/Nov, 1300 受訪者 Click to edit Master title style 11 Tame operational complexity 馴 服 運 營 的 複 雜 性 11 Click to edit Master title style 12 馴服 Kubernetes Day2 Ops 複雜性 12 • A

has been a race to create deeper networks with an ever larger number of parameters and increased complexity. In Computer Vision, several model architectures such as VGGNet, Inception, ResNet etc. (refer classification. Each new breakthrough in neural networks has led to an increase in the network complexity, number of parameters, the amount of training resources required to train the network, prediction

software-defined infrastructure platform that is built to reduce operational and infrastructure complexity. PowerFlex empowers organizations to move faster by delivering flexibility, elasticity, and simplicity containers. It works on bare-metal and virtualized servers. RKE solves the problem of installation complexity, a common issue in the Kubernetes community. With RKE, the installation and operation of Kubernetes

techniques using the metrics relevant to our use case. In some cases, these techniques can help reduce complexity and improve generalization. Let us consider an arbitrary neural network layer. We can abstract B, C, and D are all matrices. The number of MACs in a model is another metric to measure its complexity. And since they are such a fundamental block of models, they have been optimized both in hardware