RATFAT: ReAl-Time FAT for Cooperative Multitasking Environments in - - PowerPoint PPT Presentation

▶

Nov 03, 2022 172 likes •403 views

Platzhalter fr Bild, Bild auf Titelfolie hinter das Logo einsetzen RATFAT: ReAl-Time FAT for Cooperative Multitasking Environments in WSNs Sebastian Schildt, Wolf-Bastian Pttner, Felix Bsching, Lars Wolf PWSN, Boston - May 23, 2013 RAT +

SLIDE 1

Platzhalter für Bild, Bild auf Titelfolie hinter das Logo einsetzen Sebastian Schildt, Wolf-Bastian Pöttner, Felix Büsching, Lars Wolf PWSN, Boston - May 23, 2013

RATFAT: ReAl-Time FAT for Cooperative Multitasking Environments in WSNs

SLIDE 2

Wolf-Bastian Pöttner | RATFAT | Page 2 / 19

RAT + FAT?

+ ?

http://crunchall.com/wp-content/ uploads/2013/03/podgana.jpg http://5plus2.info/images/article/ article_small_33.jpg

SLIDE 3

Wolf-Bastian Pöttner | RATFAT | Page 3 / 19

Motivation

Many WSN apps need persistent data storage
Some WSN apps want large-scale data storage
Often data is stored on nodes and processed on PC
Some WSN apps require real-time guarantees

We need a file system for WSN nodes that is “PC-compatible” and real-time capable!

SD Cards

SLIDE 4

Wolf-Bastian Pöttner | RATFAT | Page 4 / 19

State of the Art in WSN file systems

COFFEE in Contiki

Unpredictable timing behavior (non real-time)
Bad performance on large (> a few MB) volumes
Not readable on standard PC Operating Systems
But: high memory efficiency

FatFs + tinyFAT

Unpredictable timing behavior (non real-time)

SLIDE 5

Wolf-Bastian Pöttner | RATFAT | Page 5 / 19

State of the Art in WSN Operating Systems

TinyOS 2

Cooperative multitasking; no preemptions
Round-robin scheduler

Contiki

Cooperative multitasking; no preemptions
Round-robin scheduler
RTimers allow (certain) real-time tasks
Interrupting ongoing SPI operations may lead to undefined states
Concurrent access to (different) SPI devices is problematic

SLIDE 6

Wolf-Bastian Pöttner | RATFAT | Page 6 / 19

Organization of a FAT Volume

#1 #2

Master Boot Record File Allocation Tables Root Directory Cluster Clusters SD Card

SLIDE 7

Wolf-Bastian Pöttner | RATFAT | Page 7 / 19

RATFAT Features

Highlights

Fully compliant to FAT specification
Support for FAT16 and FAT32 volumes and 8.3 naming scheme
Runs on any* underlying flash memory (SD, SPI flash, etc.)
Single-Buffer design (512 Byte)
Compatible to Contiki File System API
Extended API for additional functionality

Limitations

No support for „long filenames“
Keeping FATs in sync is “expensive”

SLIDE 8

Wolf-Bastian Pöttner | RATFAT | Page 8 / 19

RATFAT Architecture SPI Flash Driver SD Card driver Block IO Layer MBR Mgmt FAT16/32 Implementation Contiki CFS API FAT Extended API Application RT Management

SLIDE 9

Wolf-Bastian Pöttner | RATFAT | Page 9 / 19

RATFAT RT Overview

API

Non-blocking API
Enqueue file system requests in constant time

File System Process (FSP)

Background process handling all file system requests
Separate Stack, allows to hand control back to scheduler
Splits up FS requests into multiple atomic operations
On INGA, atomic operations take up to 12 ms
Maximum execution time per invocation
Respects system timers, yields before

SLIDE 10

Wolf-Bastian Pöttner | RATFAT | Page 10 / 19

RATFAT RT Concept

RT Process

Sampling interval T Write request Write request

RATFAT non-RT RATFAT RT

reschedule FSP reschedule FSP Deadline Miss

Write request

Deadline Miss

SLIDE 11

Wolf-Bastian Pöttner | RATFAT | Page 11 / 19

Behavior of the file system process (FSP)

                        

SLIDE 12

Wolf-Bastian Pöttner | RATFAT | Page 12 / 19

Evaluation Setup

INGA Sensor Node

Atmel ATmega 1284P: 128kb ROM, 16kb RAM

2GB Transcend Micro SD Card

Micro SD Card Slot

SLIDE 13

Wolf-Bastian Pöttner | RATFAT | Page 13 / 19

Baseline: Writing n times 8 byte with COFFEE

2000 4000 6000 8000 10000 12000 14000 1000 2000 3000 4000 5000 6000 20000 40000 60000 80000 100000 120000 140000 time per write [ms] time cumulative [ms] n'th write Average write time Median write time Maximum write time : 21.5 ms : 19.1 ms : 5573.3 ms Writing time for 8 byte Cumulative writing time

SLIDE 14

Wolf-Bastian Pöttner | RATFAT | Page 14 / 19

Writing n times 8 byte with RATFAT (non-RT)

10 20 30 40 50 60 70 80 90 1000 2000 3000 4000 5000 6000 500 1000 1500 2000 2500 3000 3500 4000 4500 time per write [ms] time cumulative [ms] n'th write Average write time Median write time Maximum write time : 0.72 ms : 0.38 ms : 71.3ms Writing time for 8 byte Cumulative writing time

SLIDE 15

Wolf-Bastian Pöttner | RATFAT | Page 15 / 19

RATFAT RT Measurement Setup

RT Process

Sampling interval T Write request Write request

RATFAT non-RT RATFAT RT

reschedule FSP reschedule FSP Deadline Miss

Write request

Deadline Miss

SLIDE 16

Wolf-Bastian Pöttner | RATFAT | Page 16 / 19

Maximum Sampling rate with RATFAT RT and non-RT

1 2 3 4 5 10 20 30 40 50 60 70 80 missed deadlines sampling rate [Hz] RATFAT-no-rt RATFAT-rt

SLIDE 17

Wolf-Bastian Pöttner | RATFAT | Page 17 / 19

Comparison of COFFEE, RATFAT non-RT and RATFAT RT

2000 4000 6000 8000 10000 1000 2000 3000 4000 5000 6000 time cummulative [ms] n'th write COFFEE RATFAT-no-rt RATFAT-rt total RATFAT-rt queueing

SLIDE 18

Wolf-Bastian Pöttner | RATFAT | Page 18 / 19

Memory Footprint Comparision

2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 COFFEE RATFAT non-RT RATFAT RT Bytes ROM (.text) RAM (.data & .bss)

INGA: 128kb ROM, 16kb RAM

SLIDE 19

Wolf-Bastian Pöttner | RATFAT | Page 19 / 19

Conclusion

Motivation

FAT on SD cards allows exchanging data between node and PC
Some applications require real-time operations on WSN nodes

RATFAT

FAT implementation for Contiki-based sensor nodes
(Optional) real-time support for cooperative multitasking env.

Results

RATFAT shows more predictable timing behavior than COFFEE
Supports real-time tasks up to 64 Hz invocation frequency

Sebastian Schildt schildt@ibr.cs.tu-bs.de Wolf-Bastian Pöttner poettner@ibr.cs.tu-bs.de

Code: http://goo.gl/Ky1Ns

SLIDE 20

Wolf-Bastian Pöttner | RATFAT | Page 20 / 19

Maximum Sampling rate with RATFAT RT and non-RT

20 40 60 80 100 120 140 160 180 20 40 60 80 100 120 140 missed deadlines sampling rate [Hz] RATFAT-no-rt RATFAT-rt

SLIDE 21

Wolf-Bastian Pöttner | RATFAT | Page 21 / 19

RATFAT Memory Footprint

ROM (.text) RAM (.data & .bss) COFFEE 10,748 Bytes 233 Bytes RATFAT non-RT 14,772 Bytes 907 Bytes RATFAT RT 19,786 Bytes 1,556 Bytes INGA (Atmega 1284p) 128,000 Bytes 16,000 Bytes

Measured on INGA (8 bit MCU)