The relationship between interstellar file system and Ethereum

2.

IPFs is a point-to-point distributed file system, dedicated to replace http

the difference between IPFs and HTTP

Security: http is centralized, and all traffic is directly carried on the centralized server, which carries great pressure and is easy to cause system crash. HTTP is also vulnerable to DDoS attacks; The storage mode of IPFs is decentralized and partitioned distributed storage, which can not be attacked by hackers, files are not easy to be lost, and security is guaranteed

efficiency: http relies on the centralized service network, so the server is easy to be shut down and the files on the server are easy to be deleted. The server needs to be turned on 24 hours; IPFs adopts P2P network topology, all computers in the whole network can become storage nodes, and the nearby distributed storage greatly improves the network efficiency

cost: http centralized server needs high maintenance cost. Once the centralized database is attacked by DDoS or damaged by force majeure, all data will be lost; IPFs greatly reces the cost of server storage and bandwidth

most of HTTP customers' network access is not localized, and there is network delay. IPFs can greatly accelerate the speed of network access, and the sense of experience will be significantly improved with the localization of network access

usage scenarios of IPFs:

< UL >

as a mounted global file system, mounted in IPFs and IPNs

as a mounted personal synchronization folder, version management and publishing are automatically performed, As an encrypted file or data sharing system, as the version package manager of all software, as the root file system of virtual machine, as the boot file system of VM, as a database, applications can write data directly to Merkle da In G data model, obtain all versions, buffers, and IPFs provided allocation

as a linked (and encrypted) communication platform

as a CDN for integrity check of large files (without SSL)

as an encrypted CDN

on Web pages, As a web CDN

as a link, there will always be a new and eternal web

< / UL >

3. Friends, computers appear: memory can not be read, the reasons to sum up, there are the following aspects
(original answer, original author: Liwang History)
1. There is a Trojan horse or virus in the computer
try: antivirus software, 360 security guard + 360 antivirus al engine version, or Jinshan guard + Jinshan drug bully,
suggestion: use "Trojan cloud killing" and "360 antivirus", "full scan" and "custom scan" viruses and Trojans, after deleting, restart the computer
after power on, click "isolate | restore" to find the Trojan horse and virus, and delete them completely
2. If the first method fails, download a "360 system first aid kit" or "Jinshan system first aid kit"
first "check and kill" and after checking and killing, delete "suspicious self boot items" and Trojan horse, and then restart the computer
then open the "isolation | recovery" area, find the "suspicious self boot item" and Trojan horse, and click "delete completely"
click "repair", "select all", and then click "repair"
3. The "player", or "chat software", or "IE browser", or "driver", or
"game" program you downloaded is unstable, or "the version is too old"! It is recommended to remove it, download a new one, or upgrade it
to the "latest version"
4. Software conflict, you have installed two or more similar software (such as: two players, two
QQ, or multiple browsers, multiple antivirus software, multiple online games, etc.)! They are not compatible with each other.
remove the "superfluous" one
5. Uninstall method: you can find "powerful uninstall software on the computer" in the "start" menu in the lower left corner of the computer, and find the rendant one to unload! "Strong cleaning" after unloading
or "360 security guard", "software manager", click, item 4: "software uninstall", click, find "rendant" and "similar" software uninstall! Such as: "player", click open, leave "storm", unload "express"! Such as: "download": click open, leave "Thunderbolt", unload "express" If it still doesn't work, go to the Internet to download a "read repair tool" and try to repair it
7. If you can't do it any more, restart the computer, press "F8" after power on, enter, enter, go into "safe mode", "Advanced startup options", find: "last correct configuration", and press to have a try to see the effect
8. If you can't do it any more, start the menu, run, enter CMD, enter, and enter () at the command prompt:
for% 1 in (% windir% & # 92; system32\*. DLL) do regsvr32.exe / S% 1
paste, enter until the screen scrolling stops, restart the computer
9. If you really can't, just "restore" the system or "reload the system"!

4. In order to meet the rapid growth of Google's data processing needs, Google has designed and implemented the Google File System (GFS). GFS has many of the same goals as distributed file systems in the past, such as performance, scalability, reliability, and availability. However, its design is also influenced by Google's application load and technology environment. It is mainly reflected in the following four aspects:

1. Node failure in a cluster is a normal rather than an exception. Due to the huge number of nodes involved in the operation and processing, thousands of nodes are usually used for common calculation, so there are always nodes in failure state at any time. It is necessary to monitor the dynamic operation of the system, detect errors, and integrate fault tolerance and automatic recovery system into the system through software program mole

2. The concept of file size in Google system is different from that in common file system. File size is usually measured in G bytes. In addition, the meaning of files in the file system is different from that of common files. A large file may contain a large number of small files in the common sense. Therefore, design expectations and parameters, such as I / O operation and block size, need to be reconsidered

3. The file read-write mode in Google file system is different from that in traditional file system. In Google Applications (such as search), the modification of most files does not cover the original data, but adds new data at the end of the file. Random writing to files is almost nonexistent. For the access mode of such huge files, the client loses the significance of data block cache, and the append operation becomes performance optimization and atomicity (a transaction is regarded as a program). It is either fully implemented or not implemented at all

4. Some specific operations of the file system are no longer transparent and need the assistance of the application program. The collaborative design of the application program and the file system API improves the flexibility of the whole system. For example, the requirement of GFS consistency model is relaxed, so that the design of file system is greatly simplified without increasing the burden of application. Atomic appending operation is also introced, so that when multiple clients are appending at the same time, there is no need for additional synchronization operation

in short, GFS is designed for the Google App itself. It is said that Google has deployed many GFS clusters. Some clusters have more than 1000 storage nodes and more than 300t of hard disk space, which are frequently accessed by hundreds of clients on different machines.

5. If you want to be faster, you can take Longyao road tunnel

6. Get on the train from Metro Line 3 (Yingao West Road Station), get off at station 15 (Yishan Road Station), walk about 3 minutes and 30 seconds to the transfer point, transfer to Metro Line 9 (Yishan Road Station), get off at station 5 (Qibao station)
get on the train at the starting point, walk about 4 minutes, then walk about 8 minutes to the destination, the journey is 31.6 km