from the Data Deluge? A case for file filtering in eBPF Giulia Frascaria October 28, 2020 1 The data deluge on modern storage 2 Compute node CPU Network Storage node Flash The data deluge on CPU Network Storage node Flash Data DoS in reverse! 11 Compute node CPU Network Storage node Flash Data So similar yet so different ● DoS is malicious ● Data transfer is business-critical ● We 12 So similar yet so different ● DoS is malicious ● Data transfer is business-critical ● We can blindly drop DoS 13 But could we reduce data transfer size? eBPF filter-reduce 14 Filter Reduce input

applications Data Storage • Easy to upgrade and downgrade • User data stored in home directory • Easy to migrate and replicate user data and configuration to another computer • User data saved in multiple are proprietary software which incur a license fee. Figure 1.6. Desktop Customisation Data Storage: User data is often located in multiple locations in Microsoft Windows, which can make backing up and place - the home directory. This makes the migration of data from an old computer to a new one easy, as well as keeping user specific backup data separate. 12 Chapter 1. Introducing Ubuntu Ubuntu Desktop

evaluation, and various development tools such as secPaver, to simplify security configuration. Data center Cloud native Edge HPC Embedded/Industrial … CPU: x86, ARM, RISC-V DPU NPU Media Bus Server codes (PACs): The register value is authenticated before being used as a pointer to access the data or code, so as to defend against return-oriented programming (ROP) and jump-oriented programming incurred in managing the huge page memory. eXpress Data Path (XDP): An eBPF-based high performance, programmable network data path that processes data before network packets enter the network protocol

packets: It breaks the limitation of the TCP framework. This model adds an EDT timestamp to each data packet based on the scheduling policy, avoiding the delays caused by large queue buffers and bringing improvement in TCP performance. 9. MultiPath TCP: It improves performance and reliability in mobile and data scenarios. MultiPath TCP supports concurrent transmission of multiple substreams in load balancing sequential write performance of large solid-state drive (SSD) blocks and the performance of the double data rate (DDR) persistent memory. 12. io_uring: This is a new asynchronous I/O framework that supports

gigabytes Perform a system back up Before installing Ubuntu Server Edition you should make sure all data on the system is backed up. If this is not the first time an operating system has been installed storage configuration screen will install Ubuntu onto the selected disk, replacing any partitions or data already there. You can choose whether or not to set up LVM, and if you do, whether or not to encrypt Type 'exit' to return. You may run 'pb-sos' to gather diagnostic data Notice: In case one needs to gather system details and diagnostic data for IBM support, this can be done here by running ‘pb-sos’ (see

wildcards for the kubernetes block like this: kubectl -n kube-system edit cm coredns [...] apiVersion: v1 data: Corefile: | .:53 { errors health kubernetes cluster.local in-addr.arpa configuration: apiVersion: v1 kind: ConfigMap metadata: name: cni-configuration namespace: cilium data: cni-config: |- { "cniVersion": "0.3.0", "name": "azure", "plugins": [ configuration: apiVersion: v1 kind: ConfigMap metadata: name: cni-configuration namespace: cilium data: cni-config: |- { "cniVersion": "0.3.0", "name": "azure", "plugins": [

Cilium Hubble Important common packages Debugging toFQDNs and DNS Debugging Mutexes / Locks and Data Races Hubble Bumping the vendored Cilium dependency Release Management Organization Release tracking Cilium’s eBPF implementation is optimized for maximum performance, can be attached to XDP (eXpress Data Path), and supports direct server return (DSR) as well as Maglev consistent hashing if the load balancing open http://localhost:12000/ to access the UI. Hubble UI is not the only way to get access to Hubble data. A command line tool, the Hubble CLI, is also available. It can be installed by following the instructions

Cilium Hubble Important common packages Debugging toFQDNs and DNS Debugging Mutexes / Locks and Data Races Hubble Bumping the vendored Cilium dependency Release Management Organization Release tracking Cilium’s eBPF implementation is optimized for maximum performance, can be attached to XDP (eXpress Data Path), and supports direct server return (DSR) as well as Maglev consistent hashing if the load balancing run in standalone mode or as a cluster making it a great choice for local testing with multi-node data paths. Agent nodes are joined to the master node using a node-token which can be found on the master

Cilium Hubble Important common packages Debugging toFQDNs and DNS Debugging Mutexes / Locks and Data Races Hubble Bumping the vendored Cilium dependency Documentation Style Header Titles Body Code Cilium’s eBPF implementation is optimized for maximum performance, can be attached to XDP (eXpress Data Path), and supports direct server return (DSR) as well as Maglev consistent hashing if the load balancing run in standalone mode or as a cluster making it a great choice for local testing with multi-node data paths. Agent nodes are joined to the master node using a node-token which can be found on the master

run in standalone mode or as a cluster making it a great choice for local testing with multi-node data paths. Agent nodes are joined to the master node using a node-token which can be found on the master wildcards for the kubernetes block like this: kubectl -n kube-system edit cm coredns [...] apiVersion: v1 data: Corefile: | .:53 { errors health kubernetes cluster.local in-addr.arpa configuration: apiVersion: v1 kind: ConfigMap metadata: name: cni-configuration namespace: cilium data: cni-config: |- { "cniVersion": "0.3.0", "name": "azure", "plugins": [